Prioritization of Roads - ReCAP - BUET - Draft Final Report - V1

Planning and Prioritisation of Rural
Roads in Bangladesh
Draft Report – Volume 1
(30 August, 2017)
Department of Urban and Regional Planning (DURP)
Bangladesh University of Engineering and Technology

(BUET)
BAN2072A
August 2017
Planning and Prioritisation of Rural Roads in Bangladesh
The analyses presented and views expressed in this report are those of the authors and
they do not necessarily reflect the views of the Government of Bangladesh, Local
Government Engineering Department, Research for Community Access Partnership
(ReCAP) or Bangladesh University of Engineering and Technology (BUET).
Cover Photo:
Mr. Md. Mashrur Rahman using
LGED’s GIS Database
Quality assurance and review table

Version Author(s) Reviewer(s) Date
1 Department URP, BUET Les Sampson and August 30, 2017
Maysam Abedin, ReCAP
ReCAP Project Management Unit

Cardno Emerging Market (UK) Ltd
Oxford House, Oxford Road
Thame
OX9 2AH
United Kingdom
Page 2
Key words
Bangladesh, Rural Road, Rural Road Prioritisation, Rural Road Network Planning, Core Road
Network, Multi Criteria Analysis, Cost Benefit Analysis, Local Government Engineering
Department.
RESEACH FOR COMMUNITY ACCESS PARTNERSHIP (ReCAP)

Safe and sustainable transport for rural communities
ReCAP is a research programme, funded by UK Aid, with the aim of

promoting safe and sustainable transport for rural communities in Africa and
Asia. ReCAP comprises the Africa Community Access Partnership (AfCAP) and
the Asia Community Access Partnership (AsCAP). These partnerships support
knowledge sharing between participating countries in order to enhance the
uptake of low cost, proven solutions for rural access that maximise the use of
local resources. The ReCAP programme is managed by Cardno Emerging
Markets (UK) Ltd.
See www.research4cap.org
Page 3
Acknowledgements
We are grateful to the Local Government Engineering Department (LGED) for providing us
access to their databases, relevant documents and necessary data required for undertaking
this study. We sincerely thank the Working Group members for this study and LGED officials
at the HQs and field offices in Tangail for their advice, guidance, help and active support in
developing the methodology, computer programme and organising the stakeholders
meetings and local level workshops in Tangail. We would also like to thank the elected
representatives of the upazila and union parishads in Tangail for their kind cooperation and
active participation at stakeholders meetings and local workshops. We gratefully
acknowledge ReCAP for their support and providing financial resources required for
undertaking this study.
Page 4
Acronyms, Units and Currencies
AADT Annual Average Daily Traffic

AHP Analytic Hierarch Process
APS Agricultural Product Surplus
BC Bituminous Concrete
BDT Bangladesh Taka
BRTC Bureau of Research Testing and Consultancy (at BUET)
BUET Bangladesh University of Engineering and Technology
CBA Cost Benefit Analysis
CC Cement Concrete
CVD Commercial Vehicles per Day
DURP Department of Urban and Regional Planning
EIRR Economic Internal Rate of Return
FRB Feeder Road Type B
GBP Great Britain Pound
GIS Geographical Information System
GoB Government of Bangladesh
HBB Herring Bone Bond (brick pavement)
HDM Highway Design and Maintenance Model
HQ Head Quarters
LGED Local Government and Engineering Department
LGI Local Government Institution
IRAP Integrated Rural Accessibility Planning
IRI International Roughness Index
MCA Multi-Criteria Analysis
MST Minimal Spanning Tree
NH National Highway
NMT Non-motorised Transport
NPV Net Present Value
PCU Passenger Car Unit
PHF Peak Hour Factor
PMGSY Prime Ministers Rural Roads Scheme (in India)
RCC Reinforced Cement Concrete
RDBMS Road Database Management System
ReCAP Research for Community Access Partnership
RED Roads Economic Decision Model
RHD Roads and Highways Department
RPPM Rural Road Planning and Prioritisation Model
RR1 Rural Road Type 1
SCF Shadow Price Conversion Factor
SFYP Seventh Five Year Plan
TTCS Travel Time Cost Saving
UCS User Cost Saving
VOCS Vehicle Operating Cost Saving
VPD Vehicles per day
WG Working Group
Page 5
Executive summary
Page 6
Contents
Key words 3
Acknowledgements 4
Acronyms, Units and Currencies 5
Executive summary ............................................................................................................9
1 Introduction ................................................................................................................. 15
1.1 Background and context of the study 15
1.2 Objectives of the research study 19
1.3 Scope of work 19
1.4 Implementation arrangements and quality assurance 20
1.5 Deliverables and milestones 21
1.6 Organisation and contents of the report 23
2 Review of planning and prioritisation methodologies in Bangladesh .
and similar other developing countries ................................................................................. 24
2.1 Introduction 24
2.2 Broad categories of planning and prioritisation methodologies 24
2.3 Prioritisation of rural roads in Bangladesh: Current practice 28
2.4 Rural road development programme in some Asian countries 30
2.5 Observations and conclusions related to developing a methodology
for the present study 32
3 Methodology for the Study ........................................................................................... 34
3.1 Introduction 34
3.2 Road categoriesand types of road development 35
3.3 Major components of the methodology for the study 36
3.4 Network planning - Defining the core upazila and union road networks and
identification of priority road development needs 37
3.5 Prioritisation of rural roads development 39
3.6 Multi-Criteria Analysis 42
3.7 Cost Benefit Analysis 49
3.8 Salient features of the methodology 53
3.9 Conclusion 54
4 Tangail District: The study area ..................................................................................... 55
4.1 An overview of the district 55
4.2 Growth centres, educational and other facilities in Tangail district 57
4.3 Rural roads in Tangail 59
4.4 Traffic volume 62
5 Local level workshops ................................................................................................... 66
5.1 Introduction 66
5.2 Objectives of the local level workshops 66
5.3 Organisation of the local level workshops 66
5.4 Participants of the local level workshops 67
5.5 Activities undertaken at local level workshops 67
5.6 Workshop materials distributed 68
5.7 Logistics needed 69
5.8 Outcome of the Local Level Workshop 69
Page7
5.9 Challenges Faced 69
5.10 Post workshop activities 70
5.11 Summary of local level workshops held in the Tangail district 70
6 Rural road planning and prioritisation software.....................Error! Bookmark not defined.
6.1 Introduction 71
6.2 Structure of RPPM 71
6.3 Data requirements 75
6.4 Description of the software 76
6.5 Programme initialisation 76
6.6 User interface 79
6.7 Data preparation and data updating for RPPM 94
6.8 Conclusion 96
7 Discussion on outputs generated……………………………………………………………………… 97

7.1 Introduction 97
7.2 Core road network 97
7.3 Improvement of roads 98
7.4 Maintenance of roads 100
7.5 Further improvements of roads 107
7.6 Conclusion 107
8 Summary, conclusions and recommendations………………………………………………………..109
References......................................................................................................................... 111
Annexures Volume 2
Executive summary
Page8
Introduction
Bangladesh has arural road network of 321,462 km with three different types of roads namely
Upazila Road, Union Road and Village Road. The Local Government Engineering Department (LGED)
in collaboration with Local Government Institutions (LGIs) manage all rural roads which connect
some 87,000 villages of the country with Zila, Upazila and Union HQs and other important market
centres. The rural roads also connect to the National and Regional Highways and Zila roads under
the Roads and Highways Department.
Bangladesh has a relatively high density of rural roads of about 2.18 km. per sq. km. However, the
major part of the network (about 70%) comprises of undeveloped earthen roads. Some important
upazila and union roads or parts of them are still unpaved. There are some 25,000 km of such roads.
Also, many villages are yet to be connected by an all-weather road.
Although every year LGED spends a considerable amount of resources for the improvement of rural
roads, the selection of roads from a large number of candidate roads is not undertaken through a
rational selection process. Consequently, many roads of higher importance may often remain
neglected while roads of lower importance are allocated resources for development. Therefore, for
the effective utilisation of available resources, there is a need to develop a planning and
prioritisation methodology for the development, upgradation and maintenance of rural roads. Such
a methodology can enhance the investment decision-making capacity in LGED andalso facilitate the
updating process of the Upazila Road Master Plan.
The Government of Bangladesh (GOB) has established priorities and strategies for rural road
development. The Seventh Five Year Plan of Bangladesh (2016-2020) has outlined priorities for rural
road development (Planning Commission, 2015, pp. 391-394). These priorities, among others,
include:
 To double lane/Upgrade and maintain selected busier Upazila Roads, Union Roads having by
a large number of commercial vehicles. At the same time, connection with rural roads with
railway and waterways will be given priority in order to promote and integrate multimodal
transport system;
 To improve Upazila Roads, Union Roads and prioritised Village Roads which have strategic
importance to connect road network, railway and waterways.
The Department of Urban and Regional Planning (DURP) at Bangladesh University of Engineering and
Technology (BUET) in collaboration with LGED, is implementing a project to develop an appropriate
planning and prioritisation methodology for the development of rural roads in Bangladesh. The
project (Project reference: BAN2072A) is being sponsored by the Research for Community Access
Partnership (ReCAP) programme, a UK aid funded six-year programme, with an overall aim to
promote safe and sustainable rural access in Asia and Africa.
The main objectives of the project, among other things, include:

 Development of a methodology on planning and prioritisation of rural roads for their
development and maintenance
 Development of a software to implement the developed methodology, and
 Production of a user manual for the software and training of 15 professionals to use the
software.
Page9
This reports provides, the details of the methodology developed for the purpose of this study and
the software for its implementation. It may be noted that the production of a user manual for the
software and training of 15 officials are not within the scope of this report. These activities will be
covered in a separate report after their completion in the following months.
Methodology
LGED has already developed a network of about 74,000 km of paved roads and about 20,000 km of
brick paved roads which may be paved in the near future. Any further development of roads should
be selective and based on some criteria so that the network can be sustained and properly
maintained over a longer term, and enhance rural access. The selection criteria, among others, may
include considerations of access to all villages and connectivity with higher order centres and roads.
Some broad criteria, similar to national programmes in other countries, can be considered to
develop core networks of rural roads at upazila and union levels which can provide rural accessibility
at all levels as well as ensure intra- and inter-upazila connectivity. Fortunately, such national level
broad policies are already in place that have been considered to develop such core networks of rural
roads.
In developing the methodology the following matters have been considered:

 Government objectives, current policies and strategies as stated in official documents
 Guidance and advice of the Working Group established for the project
 Comments and suggestions received at the First Stakeholders Workshop
 Outcome of consultation meetings with concerned LGED officials at the HQ and field offices
and current practices of LGED
 Observations and conclusions distilled from the literature review
Following discussion with LGED officials it was decided that methodologies on prioritisation would
be developed for three types of road development works namely, improvement, further
improvement (upgradation of an existing paved road) and maintenance of rural roads. The
definitions of these three terms are as follows:
Improvement
 Converting an earthen road to a paved road i.e., from earthen to BC/RCC in an existing
alignment
 Converting a partly paved road to a fully paved road
 Converting an HBB road to a fully paved road
Further Improvement/Upgradation
 Improvement of road geometric standards, raising of embankment and widening of
pavement and/or road crest and raising of road embankments of an existing road
Maintenance
 Maintenance of an already paved or partly paved road (BC, RCC or HBB).
The methodology developed for this study has two major components. The first component follows
a network approach to road planning; and the second component involves prioritisation of road
development based on evaluations using a set of physical, economic, social and other criteria (see
Figure 3.2 in chapter 3). Both the components include the provision of local stakeholders’
participation in the process.
Page10
All Upazila and Union roads, some important Village roadsand national roads under Roads and
Highways Department (RHD), if any, form the core network in an Upazila. In the prioritisation
scheme all such roads have been given additional importance as together they form a network vital
to maintain inter- and intra-Upazila connectivity. The activity centres form the nodes and Upazila
andUnion roads and selected village roads connecting them are the links of the core network.
The organisation of local workshops is an important element of the methodology. Local stakeholders
validate the core network generated by the software according to the defined criteria of core
network. The stakeholders also define local priorities for the development and maintenance of
roads, identify potential upazila and union roads that may need further development, and evaluate
the connectivity status of the roads in the core networks.
The methodology considers evaluation of all rural roads in an upazila to determine the priority for
their development and/or maintenance. Maintenance is considered only for paved roads. Two
evaluation methodologies namely, cost benefit analysis (CBA) and multi-criteria analysis (MCA) are
used for the evaluation of roads. For maintenance, only MCA has been applied, however. The
priority ranking of roads are based on the outcome of evaluations made by applying CBA and/or
MCA.
The benefits of improving rural roads may be estimated in terms of savings in vehicle operating costs
and travel time costs. A simplified CBA methodology has been developed to estimate such benefits
considering the availability of data mainly from the databases of LGED, and data from LGED and
other secondary sources. The adopted methodology, however, follows the standard procedures in
estimating economic costs and benefits of road improvement as far as practical. In developing the
CBA methodology, methodologies developed for other recent studies on rural roads including a CBA
model developed in-house by LGED were considered. The developed CBA model for the study
estimates EIRR values for all roads that require improvement or further improvement as defined
earlier.
Prioritisation methodologies should not be based solely on the traditional economic evaluation of
roads. Other important criteria such as, connectivity with higher order roads and benefits from
enhanced access to socio-economic infrastructures should also be considered. Further, rural roads
provide access benefits to people living along the roads many of whom may not be using any
vehicular traffic. Therefore, in order to make a more complete assessment, the access benefits to
people also need to be assessed by an MCA or some other methodology. For MCA, criteria such as
traffic volume, number and type of socio-economic facilities, growth centre and rural market served
by a road, connectivity and local priority have been considered. The applicable criteria for MCA were
selected from among a list of potential criteria by conducting an Analytic Hierarchy Process (AHP)
survey among the stakeholders including experts and LGED officials.
After selection of the MCA criteria, their relative weights were established. The AHP technique,
which is commonly used for the establishment of relative weights of selected criteria, has been used.
A survey instrument was designed and a survey was conducted among the experts and LGED officials
at the HQ and field offices and other stakeholders to determine the relative weights of the selected
criteria for their use in MCAs.
As the actual values of the selected criteria are in different units of measurement and the range of
their minimum and maximum values widely vary, the actual values were standardised on a common
scale of 0 to 100 to allow for their summation. Finally, the priority score or index value of each road
was calculated by summing the product of the standardised value of each criterion by its relative
weight as follows:
Page11
I   ( wi  xi ) , where
I= Combined score, wi= Relative weight of criteria i, xi= Standardised score of criteria i
The priority order for the development of individual roads is determined on the basis of their
combined scores.
Some socio-economic facilities are of the same type but do not have the same order of importance.
For example, different types of educational institutions such as primary school, high school, college
and madrasa. For these types of facilities, the Reed-Muench Median Threshold Population Method
was applied to determine their relative weights.
The following table shows the selected criteria and their AHP weights for each type of road
development.
MCA criteria Improvement Improvement of Further Maintenance of

of unpaved partly paved and improvement of roads
roads HBB roads roads
Traffic Volume 7.95 20.01 21.84 15.05
Facilities Served 18.65 15.02 9.54 14.85
Growth Centre/ 29.20 13.58 17.91 12.70
Market Served
Connectivity 35.80 21.04 20.07 19.65
Local Priority 8.40 5.97 14.93 N/A
Surface Type N/A 5.86 N/A 7.80
Road Type N/A 11.83 7.86 14.15
Road Safety N/A 6.69 7.84 N/A
Bus Route N/A N/A N/A 11.50
Last maintenance year N/A N/A N/A 4.30
Total 100.00 100.00 100.00 100.00
Notes: All values are expressed in percent. N/A = Not applicable. A full list of the facilities and their relative
weights determined by the Reed-Muench Median Threshold Population Method are provided in Table 3.10 in
section 3.
The developed methodology for this study has the following important features. It considers
government strategies and policy objectives as outlined in official policy documents. These policy
objectives focussing on rural access have been used to generate core road networks at upazila and
union levels. The prioritisation of roads in the core networks can enhance rural access at all levels,
and ensure intra- and inter-upazila accessibility.
The methodology combines the top-down and bottom approaches to planning. The participation of
stakeholders at local workshops for considering their inputs to prioritise rural road development is
an in-built feature. It also fulfils the strong demand of local leaders expressed at the first stakeholder
workshop.
Page12
LGED undertakes three types of rural road development activities: improvement, further
improvement and maintenance. All of these activities have been considered within a single planning
framework based on a consistent approach to evaluation of priority.
It was decided in consultation with LGED that the developed methodology would be applied in a
pilot district. Tangail was selected as the pilot district. The selection of the pilot district considered,
among other factors, representativeness of the district, likely disruption of field work due to natural
events and other matters. However, the methodologies developed were generalised, as much as
possible, to be applicable to all over Bangladesh.
Accordingly, the methodology was applied in the 12 upazilas of Tangail. In collaboration with LGED,
local workshops were organised in each of these 12 upazilas to facilitate participation of local
stakeholders in the planning process, and collection of required inputs from them.
The planning and prioritisation software
A new software was required to implement the developed methodology. Customisation of any
existing software/application was not possible to implement the methodology. Accordingly, the
project team has developed a road network planning and prioritisation software, the Rural Road
Planning and Prioritisation Model (RPPM), to meet the road development requirements of LGED.
RPPM is a web-based tool designed to enhance planning and decision making capacity in LGED.
RPPM has been implemented on the GIS portal of LGED and can be used by LGED officials working at
its HQs or field offices. Its integration with the GIS portal facilitates mapping and generating road
priority outputs on the web. The software can generate core road networks at upazila and union
levels, and evaluate the priority of rural roads for their development including their ranking based on
the outcome of MCA and CBA analyses.
As requested by LGED, RPPM can produce eight separate priority lists for different surface types of
roads and different types of development. These categories are listed in the following table.
Priority list Type of road development Type Surface type

1 Low traffic volume earthen road
2 Improvement High traffic volume earthen road
3 Partly paved road
4 Fully HBB road
5 Partly paved road
6 Maintenance Fully HBB road
7 Fully paved road
8 Further Improvement Fully paved (Full BC/RCC, HBB + BC/RCC)
RPPM can also generate three other types of output:

 Core networks at upazila and union levels
 Priority score table with basic information, details of priority scores, and CBA and MCA
results of a road, and
 Maps showing prioritised roads on an upazila map.
The priority score table for a road have the following information:
 Road code and road name
Page13
 Roughness condition, AADT and CVD values
 Total road length, road type, surface type, connectivity
 MCA scores for facilities
 Evaluation score including EIRR and MCA scores
 Priority ranking considering EIRR or MCA.
For road maintenance, only MCA scores and ranking by MCA scores are shown. It may be mentioned
here that road maintenance is considered only for paved roads. As CBA, factoring regular and
periodic maintenance costs, would be undertaken before an unpaved or partly paved or HBB road is
considered for improvement in the first place, the maintenance priority of the road is based on its
MCA score.
For the pilot district, RPPM calculated MCA scores and EIRR values as expected. However, CBA for
most of the low volume village roads did not produce any meaningful results. In fact, for many low
volume roads it was not possible to calculate the low (negative) EIRR values. As such, the ranking of
low volume roads for improvement is based on their MCA scores.
RPPM uses data mainly from the databases of LGED. Two new tables have been added to the road
database for use by RPPM. The first table contains information on connectivity of roads, local
priority order and road safety. These information are required for MCA. The second table contains
values of VOC, TTS, speed, average occupancy and PCU values by vehicle type, which are required
for CBA. The information in the first table are collected from local workshop of stakeholders. The
information in the second table comes from a study undertaken by GTZ for LGED. When results from
a more recent study are available, the values in this table may be replaced.
Before running the software for another district, these two additional tables have to be added to the
road database for that district. Some parameter values used in CBA such as unit construction and
maintenance costs, shadow price factor, tax rate etc. may change over time. Unit costs may also vary
for other districts. As such, these default values may be reviewed before running the software for
another district and, if necessary, may be changed by the user.
The quality of output of any software depends on the quality of input data. It is recommended that
before applying the methodology in other districts, the road database, especially data on traffic
volume may be updated. Both the CBA and MCA analyses use traffic data. CBA results, as expected,
are sensitive to traffic volume data.LGED may also consider updating the database periodically.
LGED maintains two databases: the road database and the GIS database. These two databases are
not fully integrated. If an integrated database were available, the wealth of information available in
these two databases could be effectively used in undertaking many useful interactive analyses
relevant to rural road development and other development purposes. Many useful analyses would
be possible even from within the GIS platform without requiring a custom-made program. Some
other government departments could also take the benefit of such analyses in preparing their
development plans. Although it may require considerable amount of time, resources and efforts to
integrate the two databases, it would be worth considering such investment given its potential use
in development planning.
Page14
1 Introduction
1.1. Background and context of the study
The majority of about 162 million people of Bangladesh lives in rural areas. There are some 87,000
villages spread over an area of about 147,500 sq.km. Bangladesh has 64 Zilas (districts), 489 Upazilas
(sub-districts) and 4,553 Unions. A rural road network of 321,462 km withthree typesof roads
connects the villages/settlements with Zila, Upazila and Union HQs and other important market
centres as well as national and regional highways and Zila roads.
The Local Government Engineering Department (LGED) in collaboration with Local Government
Institutions (LGIs) manage all rural roads which form more than 93% of total roads in the country
(of about 342,764 km). The Roads and Highways Department (RHD), the national highway authority
of Bangladesh, manages the National and Regional Highways and Zila (District) roads. The different
types of roads and their responsible agencies are given in Table 1.1.
Table 1.1: Types of Road and their responsible authority

Road type Definition Responsible
authority
National Highway Highways connecting National capital with RHD
Divisional HQs or seaports or land ports or Asian
Highway.
Regional Highway Highways connecting District HQs or main river or RHD
land ports or with each other not connected by
National Highways.
Zila (District) Road Roads connecting District HQs with Upazila HQs or RHD
connecting one Upazila HQ to another Upazila HQ
by a single main connection with National/
Regional Highway, through shortest
distance/route.
Upazila (Sub-district) Roads connecting Upazila HQs with Growth LGED & Local
Road Centres or one Growth Centre with another Government
Growth Centre by a single main connection or Institutions (LGI)
connecting Growth Centre to Higher Road
System, through shortest distance/route.
Union Road Roads connecting Union HQ/s with Upazila HQs, LGED and LGI
growth centres or local markets or with each
other.
Village Road (Types A a) Roads connecting Villages with Union HQs, local LGED and LGI
and B) markets, farms and ghats or with each other.
b) Roads within a Village.
Source: GoB, 2005; Training Manual on Road Maintenance Management, (RIMMU), Aug 2006 as quoted in GTZ (2009)
Table 1.2 shows the status of rural roads in Bangladesh. Compared with other developing countries,
Bangladesh has a relatively high density of rural roads of about 2.18 km. per sq. km; for example, the
national average of road density in the Philippines is 0.62 km. per sq. km. However, the major part of
the network (about 70%) comprises of undeveloped earthen roads. Further, many of these unpaved
Page15
roads have ‘gaps’ where bridges/culverts are required to be built. The shares of road length by broad
types are shown in Figure 1.1.
Table 1.2: Status of rural Roads in Bangladesh (as of June 2016)

Type Number of road Road length by surface type (km) Total
Fully Fully Partly Earthen Earthen Flexible Brick Rigid No.
paved paved paved pavement paved pavement (km)
(BC) other (BC) CC/RCC
Upazila 2,653 541 1,384 186 6,848 2,7695 2,188 668 4,764
Road (37,399)
Union 1,996 780 3,909 1,321 18,384 18,480 4,125 631 8,006
Road (41,621)
VR Type A 4,297 2,567 11,347 26,480 93,021 18,517 8,422 938 44,691
(120,899
VR Type B 2,835 2,562 8,589 58,174 109,130 6,639 5,247 527 72,160
(121,543)
Total 11,781 6,450 25,229 86,161 227,383 71,332 19,982 2,764 129,621
(321,462)
Source: LGED (Internal communication with the BUET Team)
Notes: VR = Village Road; BC=Bituminous Concrete; CC=Cement Concrete; RCC: Reinforced Cement Concrete
The transport demand for passenger and goods has increased manifold over the years due to
consistent annual growth of the national economy between 5 to about 7%. In addition, with the
increase in personal income and literacy rate, the lifestyle and travel pattern of the rural people are
also changing. The increased demand for facilities like schools, colleges, clinics, hospitals, banks,
markets, places of recreation etc., is changing the economic geography of the country; many rural
settlements are transforming into small towns and small towns into medium-sized cities. People are
traveling more and using better and faster modes of transport to travel between these centres and
the villages
.
Fig 1.1(a): Share of road length in km by road type Fig 1.1(b): Share of unpaved (earthen) road
length in km by road type
Source for figures 1.1(a) and 1.1 (b): Based on data in Table 1.1
Page16
In response to increased rural transport demand, LGED under the Ministry of Local Government and
Rural Development has been implementing a massive rural road and other infrastructure
development programme since 1990s. As has been observed in The Seventh Five Year Plan (p. 391):
“Most of the Upazila/Union roads of LGED were constructed during 1990-2010 with
an objective of rapid development of rural infrastructure for economic growth. These
roads were built overold earthen embankment owned by Union Parishads or Zilla
Parishads and they do not have adequate geometric standards.”
In the course of time, the rural roads are being used by higher and heavier volume of traffic than
envisaged during their design and construction. Consequently, many rural roads, originally built for
low volume light traffic, are found inadequate and substandard to cope with the growth of traffic
volume. Substantial growth in traffic volume and change in their composition have led to rapid
deterioration of the road infrastructure. Consequently, more frequent road maintenance is required;
accidents on rural roads are also taking place with increased frequency and severity. There is a need
to upgrade the width and other geometric features as well as construction standards of rural roads.
Figure 1.2 shows the share of surface type for each category of road.
Fig 1.2(a): Share of surface type of upazila roads Fig 1.2(b): Share of Surface type of union roads
Fig 1.2(c): Share of surface type of village road Fig 1.2(d): Share of surface type of village road
type A roads type B roads
Source for figures 1.2(a), (b), (c) and (d): Based on data in Table 1.1
Page17
The Government of Bangladesh (GoB) has established priorities and strategies for rural road
development. The Seventh Five Year Plan of Bangladesh (2016-2000) has outlined priorities for rural
road development (Planning Commission, 2015, pp. 391-394). These priorities, among others,
include:
 To double lane/Upgrade and maintain selected busier Upazila Roads, Union Roads having a
large number of commercial vehicles. At the same time, connection with rural roads with
railway and waterways will be given priority in order to promote and integrate multimodal
transport system;
 To improve Upazila Roads, Union Roads and prioritised Village Roads which have strategic
importance to connect road network, railway and waterways.
The Five Year Plan also identifies major strategies as follows:
 Rural infrastructure development/ improvement will be planned and implemented based on

the findings of Effect/ Benefit/ Impact/ feasibility studies; and
 A simpler and quicker way for feasibility study and deriving the EIRR will be developed for
the rural roads. There are competing needs for various types of rural infrastructure, such as,
Upazila Roads, Union Roads, Markets, Ghat facilities, etc., and even for roads alone, there is
need for improved maintenance and bridging the gaps. At the spatial level, there are
competing needs for different geographical regions. A guideline for investment prioritisation
and selectivity will be developed and calculation of economic rate of return will be adopted
to guide the major investment decisions.
 Rural Road Master Plan of LGED will be updated. The updated Master Plan will be followed
for infrastructure development projects covering Upazila and Union roads including
bridges/culverts, bridges/culverts on village roads and development of growth
centres/markets, ghats and Union headquarters.
LGED is required to prepare a master plan for Upazila roads, Union roads and Village Roads along
with Growth Centres/Rural Markets, Union Parishad Complex, etc. In the process, the Upazila
Engineer updates the road inventory of Upazila as per necessity and sends the updated version of
the inventory of roads and other facilities, and the map to the GIS Unit at the LGED head office for
maintaining the updated GIS database at the national level.The stated main objectives of the Rural
Road Master Plan are as follows:
 To identify and prioritise a useful and effective rural road network throughout the country to
ease the rural life as a whole
 To provide all-weather access to all Growth Centres, all Union Parishad Complexes, most of
the rural markets and other service delivery centres in rural areas
 To improve rural accessibility for facilitating agricultural production and marketing of
different products
 To reduce poverty through employment generation and accelerating economic activities in
rural areas
 To strengthen the Local Government Institutions and promoting local governance by
providing technical assistance as needed.
Page18
Although LGED spends a considerable amount of resources for the improvement of rural roads, the
selection of roads from a large number of candidate roads is not undertaken through a rational
selection process. Consequently, many roads of higher importance may often remain neglected
while roads of lower importance are allocated resources for development. LGED has already
developed a network of about 74,000 km of paved roads and about 20,000 km of brick paved roads
which may be paved in the near future. Any further development of roads should be selective and
based on some criteria so that the network can be sustained and properly maintained over a longer
term, and enhance rural access at all levels.
Therefore, for the effective utilisation of available resources, there is a need to develop a planning
and prioritisation methodology for the development, upgradation and maintenance of the rural
roads. The methodologyalso can facilitate the updating process of the Upazila Road Master Plan and
the investment decisions for rural road development in Bangladesh.
The Department of Urban and Regional Planning (DURP) of Bangladesh University of Engineering and
Technology (BUET) in collaboration with LGED, is implementing a project to develop an appropriate
planning and prioritisation methodology for the development of rural roads in Bangladesh. The
project (Project reference: BAN2072A) is being sponsored by the Research for Community Access
Partnership (ReCAP) programme, a UK aid funded six-year programme, with an overall aim to
promote safe and sustainable rural access in Asia and Africa. The project formally started in April
2016. The Terms of Reference (ToR) is given in Annex A, Volume 2.
1.2. Objectives of the research study

The objectives of the study, as stated in the ToR are:
 To examine existing prioritisation methods in Bangladesh and other similar countries and
determine whether they provide optimal solutions for rural roads;
 To develop a prioritisation method, software and user manual for Bangladesh and test it in a
region for gaining full confidence; and
 To train 15 professionals on the use of the planning and prioritisation tool.
It is expected that the outcomes of the research project will enhance the decision making capacity in
LGED regarding selection of rural roads for improvement, further improvement/upgrading and
maintenance and thereby ensure proper utilisation of public money for socio-economic benefits of
the people.
1.3. Scope of work

The main purpose of this research study is to identify the rural roads in an Upazila that should be
prioritised for development to maintain rural accessibility as well as connectivity with the
surrounding upazilas and the greater region. The major outcomes of the study include:
 Develop a planning methodology for the identification of core networks at the Upazila and
Union levels that can ensure rural accessibility at all levels
 Develop priority ranking methodologies for the three types of rural road development works
undertaken by LGED namely, improvement, further improvement/upgradation and
maintenance.
Page19
 Design and develop a software to implement the above-mentioned planning and
prioritisation methodologies
 Enhance the decision making capacity in LGED regarding selection of rural roads for
development by training 15 officials in applying the methodologies and using the developed
software.
The definitions of the above-mentioned (the second bullet point above) three types of road
development are as follows:
Improvement
 Converting an earthen road to a paved road i.e., from earthen to BC/RCC in an
existing alignment
 Converting a partly paved road to fully paved road
 Converting a HBB road to fully paved road
 Improvement of road geometric standards such as raising of embankment height,
road widening and higher pavement design standards.
Maintenance
 Maintenance of an already paved road (BC/CC or HBB).
The prioritisation methodologies should not be based solely on the traditional economic evaluation
of roads. Other important criteria such as, connectivity with higher order roads and benefits from
enhanced access to socio-economic infrastructures should also be considered. Accordingly and as
required by LGED, EIRR values are to be calculated for Union, Upazila and village roads based on a
simplified cost benefit analysis methodology (CBA) considering the availability of data mainly from
databases of LGED and other secondary sources. In addition to this, prioritisation methodologies are
to be developed considering improved access to socio-economic facilities, and other factors, which
can contribute to people’s welfare.
It was decided in consultation with LGEDthat the prioritisation methodologies would be developed
considering Tangail as the pilot district. The selection of the pilot district considered, among other
factors, representativeness of the district, likely disruption of field work due to natural events and
other matters. However, the methodologies to be developed should be generalised, as much as
possible, to be applicable to all over Bangladesh.
The prioritisation models should be simple and easy to use and must not require any vast amount of
additional data collection. They should be integrated with the existing databases of LGED so that
visualisation of the prioritisation results may be possible. It is expected that the computer-based
prioritisation models will be used to enhance decision making capacity in LGED at itsHQ and district
and upazila offices. The planning and prioritisation road development software is also expected to
facilitate the updating process of the Upazila Road Master Plan.
1.4. Implementation arrangements and quality assurance

The Department of Urban and Regional Planning (DURP) in collaboration with the Departments of
Civil Engineering and Computer Science and Engineering at BUET, and LGED has implemented the
project. LGED has also provided direct support to the project team in undertaking field studies in the
pilot district.
Page20
The project (Project reference: BAN2072A) has been sponsored by the Research for Community
Access Partnership (ReCAP) programme, a UK aid funded six-year programme with an overall aim to
promote safe and sustainable rural access in Asia and Africa.
LGED established a Working Group (WG) to ensure the quality of project outputs and assist in the
implementation process. The Working Group has provided overall guidance and assistance in
implementation of the project, considered the intermediate outputs especially the proposed
methodology, and also gave necessary advice to the project team as required. The project team has
also received feedback on initial software outputs from senior officials of LGED which helped to
finalise the computer program. The formation of the WG and the list of its members are provided in
Annex B, volume 2.
1.5. Deliverables and milestones

This study started on 7 April 2017. An inception report outlining problem analysis, literature review,
proposed methodology for the study, scope of work, work plan, and implementation arrangements
was submitted on 5 May 2016. Over a period of the following six months the Project team worked
out a detail methodology of the study with guidance and advice of the Working Group, a series of
consultation meetings with concerned LGED officials at the HQ and field offices in Tangail, and
comments and suggestions received at the First Stakeholder Workshop held on 16 August 2016. The
list of meetings held in different times is provided in Annex C, volume 2.
The methodology and the flow chart for the section of roads for different types of development
were finalised at the third working Group Meeting held on 6 October 2016. However, some minor
changes to the flowchart were undertaken following meetings with LGED officials. Field works were
undertaken in the pilot district after finalisation of the methodology and the flow chart. The list of
officials met from time to time is given in Annex D, volume 2.
The Project team, with the support of LGED, organised 12 local level workshops in 12 upazilas of
Tangail. Workshops were completed in early February 2017. The workshops provided an opportunity
for the local stakeholders to be involved in the planning and prioritisation process of rural road
network development in their area as well as validate the preliminary core road network generated
based on the data available from LGED’s databases. The organisation of such local workshops is a
part of the planning and prioritisation methodology.
The BUET Team (provided in Annex E, volume 2) has been working to develop a software application
since finalisation of the methodology. The developed software, in its current form, can be used as a
web-based tool by LGED officials working at its HQs or field offices to enhance decision-making
capacity in LGED. The software can generate core road networks at upazila and union levels, and
evaluate the priority of rural roads for their development including their ranking based on the
outcome of MCA and CBA analyses.
The following are the main deliverables of the project:

 Inception report
 First Stakeholder Workshop
 Draft Report
 A software on Planning and Prioritisation of Rural Roads
 Second Stakeholder Workshop
Page21
 Final report with recommended planning/prioritisation methodology
 Production of Prototype Illustration of GIS Application
 Production of a User Manual on the planning/prioritisation methodology
 TOT training of 15 Professionals and report
Some of these deliverables are also milestones of the project. Table 1.3 provides the milestones,
their due dates and current status.
Table 1.3: Project milestones and Due Dates (as per amended contract in June 2017)
ReCAP Project Planning and Prioritisation of rural roads in Bangladesh
Title
Project Start date 7 April 2016
duration End date 31 December 2017
Item Due date (as Current
mentioned in the status/Comments
revised contract)
Inception Report 5 May 2016 Delivered on 5 May
2016;
Revised Report July
2016
Milestone First Stakeholder Workshop 16 August 2016 Delivered on 9
August 2016
Draft Report for presentation 30 August 2017 Present Report

at the Second Stakeholder
Workshop
Final Report with 29 September 2017 In progress. Will be
recommended finalised after the
planning/prioritisation Second Stakeholders
methodology and software Workshop
Production of a User Manual 15 October 2017 In progress
Production of Prototype 30 October 2017 Field work in

Illustration of GIS Application progress
Delivery of ToT Training and 30 November 2017 Preparations will be
Report initiated in October
2017
This reports provides, the details of the methodology developed for the purpose of this study and
the software for its implementation. It may be noted here that the user manual for the software and
training of 15 officials are not due yet and as such outside the scope of the present report.
Page22
1.6. Organisation and contents of the report
The report is organised in two volumes. The first volume presents the main text of the report
including an executive summary, the planning and prioritisation methodology and an overview of
the software developed to implement the methodology. There are eight chapters in this volume.
Chapter 1 provides the background and context of the study, objectives, scope of work, and
deliverables and milestones of the project. Chapter 2 provides a review of methodologies on rural
road planning. It also includes a review of road development programme in some Asian countries,
the current practice of prioritisation for road maintenance by LGED and some conclusions. The
details of the methodology developed and adopted for the study in the pilot district is presented in
Chapter 3. Chapter 4 is on rural roads in the pilot district. Chapter 5 presents discussions on field
studies and local level workshops organised in the pilot district. An overview of the software
developed to implement the methodology is presented in Chapter 6. Chapter 7 is on discussions on
the output. Chapter 8 is the last chapter and it includes conclusions, observations and some
recommendations for consideration of LGED.
The second volume contains all the appendices referred to in the first volume, reports of local
workshops in 12 upazilas of the pilot district, and outputs generated by the software for the 12
upazilas.
Page23
2 Review of planning and prioritisation methodologies in Bangladesh and
similar other developing countries
2.1. Introduction
Rural road development has been at the forefront of rural development for many decades. In the
1950s and 1960s, rural infrastructure development, particularly rural roads, was taken as the
starting point of economic development in many developing countries. In recent times, some
countries have introduced more broad-based programmes to extend basic infrastructure such as
rural roads, irrigation and water supply, soil conservation and social facilities in remote and poverty-
stricken areas. Bangladesh, India, Malaysia, Nepal, Pakistan, the Philippines, Sri Lanka and Thailand
have all introduced such programmes.
Ample empirical evidence of the positive impact of rural roads on poverty reduction as well as
improvement of socio-economic conditions of the rural people has emerged from studies conducted
by many international organisations in a number of countries in the region, including Bangladesh,
China, India, Indonesia, the Philippines, Sri Lanka, Thailand and Viet Nam. Research studies have
considered different approaches in addressing the planning, policy and methodological issues
concerning the definition, identification, selection and evaluation of rural roads. The national
programmes in countries also differ in their approaches to selection and evaluation of rural roads for
making investment decisions. This chapter presents a review of such approaches and considers some
useful considerations from such reviews to develop a planning and prioritisation methodology for
rural roads in Bangladesh.
2.2. Broad categories of planning and prioritisation methodologies

A literature review on rural road planning and prioritisation methodologies in Bangladesh and similar
other developing countries reveals that such methodologies may be grouped into five broad
categories:
1) Multi Criteria Analysis (MCA) (also known as Prioritisation Ranking or PR techniques)

2) Economic analysis (Cost Benefit analysis also referred to as Benefit Cost analysis)
3) Participatory approaches
4) Network planning models based on optimisation techniques
5) Hybrid methodologies - combination of multiple methodologies on road planning
In the context of the present study, these five broad categories are discussed next.
2.2.1. Multi Criteria Analysis (MCA)

Multi Criteria Analysis (MCA) or Prioritisation ranking (PR) techniques of rural roads have been
widely used for the selection of rural road projects in many countries of Asia and Africa. The ranking
of projects is based on a set of factors or criteria reflecting policy or social objectives of the
government. Such factors may or may not include cost variables but may include variables such as
population served, improvement of access to markets and services, impacts on natural systems,
traffic volume, social and economic welfare etc. After selection of factors, their relative weights are
established. The use of Analytical Hierarchy Process (AHP) is common for the establishment of
relative weights of the factors. Finally, the priority score or index value of each candidate road
project is calculated by summing the product of the value of each factor (often the actual value is
normalised) by its relative weight.
Page24
Many good examples on application of MCA can be found in the literature, for example, ESCAP
(1979), Greenstein and Bonjack (1983), UNCHS (1985), Lebo and Schelling (2001), Hine et al. (2003),
Bhandari et al. (2014), and Philippine Rural Development Project (2015, pp. 11-13).
It may be noted that in Bangladesh LGED uses some criteria such as classification of roads, surface
type, gaps, traffic volume (CVD), fund source, markets, hospitals, social centres, educational centres
and industry, and some relevant weights of these criteria for prioritisation of rural roads
maintenance. For example, for Upazila road a score of 12 and for Village road a score of 3 is used.
However, apparently, the selection of these criteria and assignment of their weights do not follow
any rational basis. Further details on this practice are provided in section 2.3.
The MCA methodology is used to rank rural road investments, and is typically applied when traffic
volumes are less than 50 vehicles per day and too low to apply the conventional cost benefit analysis
to make sense, but there is a strong belief that there will be important social benefits arising mainly
from improvement of access and connectivity.
It is important to note here that in addition to such benefits, there are impacts of road investment
on socio-economic development, or the so-called indirect- and induced benefits. Alternative impact
assessment methods may be used to assess such benefits. This can be supported by additional data
collection methods through which the proposed impact of rural road investments can be estimated
and measured by obtaining views from communities with respect to the expected impact on socio-
economic conditions on communities. A study by Ahmed and Hossain (1990) in Bangladesh is a good
example for such impact assessment.
It may be worth mentioning here that the Roads and Highways Department of Bangladesh has
developed a methodology for prioritisation of national and regional highways considering
importance of service centres, condition of road, and level of services required. The Project
Appraisal Framework prepared by the Planning Commission adopted a multi-criteria evaluation
framework for road projects incorporating a set of equity and efficiency criteria for the selection and
prioritisation of highway projects (Planning Commission, 2005).
Generally, indicators used in MCA implicitly reflect economic and subjective evaluations (Lebo and
Schelling, 2001). If the weights and factors are decided upon and allocated in a participatory way,
MCA has the potential to be a participatory planning method (see discussions below on participatory
approach). The MCA methodology should not be applied without consultation with the concerned
users and stakeholders. The outcome of the MCA methodology can become non-transparent,
especially if too many factors are considered and acomplicated formula applied to determine an
index value or priority score. Therefore, as suggested by Lebo and Schelling (2001), if adopted, this
method should be kept simple, transparent, and participatory.
Despite all the advantages of MCA, its main caveat is the inability to compare the benefits with
respect to the costs. In order to address this problem, Gühnemann, Laird, and Pearman (2012)
developed an integrated approach where they combined CBA and MCA for effectively evaluating
road development projects. With this approach the strengths of both methods are retained.
2.2.2. Economic analysis (Cost Benefit Analysis)

The application of Cost Benefit Analysis (CBA) has been widely adopted in analysing the merits of
road investment projects in both developed and developing countries including Bangladesh. CBA has
been popular among the decision makers as it provides a systematic framework to evaluate all costs
and benefits of a project. In short, CBA identifies and estimates in monetary terms all costs and
Page25
benefits of a project during its (economic) lifetime, discounts them to a reference point in time
(generally the present year), and compares to see if the summation of benefits exceed the
summation of costs. The commonly applied assessment criteria for project selection include, Benefit
Cost Ratio, Net Present Value (NPV), Internal Rate of Return (IRR), Payback Period, etc.
There are several approaches to estimate the benefits of road development. The User Cost Saving
(UCS), Vehicle Operating Cost Saving (VOCS) and (Travel Time Cost Saving (TTCS), and Agricultural
Product Surplus Method (APS) are among such conventional approaches. However, in recent years,
the most commonly applied CBA methodology follows the VOCS and TTCS approach. This approach
has also been applied in Bangladesh for many road infrastructural development projects (for
example, PRMIMP-II, RDP, TRIDP, RDP-24, RIIP-I and RTIP).
Due primarily to its wide application and popularity among decision makers, several computer based
CBA models have been developed to provide a harmonised approach to economic evaluation of road
projects. One of the most well-known such model is the Highway Design and Maintenance Model
(HDM) developed by the World Bank. The original model did not include the maintenance issue. The
HDM-III, introduced in 1980s, included the maintenance issue. However, this model had many
limitations especially reliance on vehicle operating cost which may not be very relevant in the
context of rural roads in developing countries.
A new model, HDM4 was developed to address the limitations of the HDM-III model. HDM4 is a
completely new software package which can be used as the primary tool for the analysis, planning,
management and appraisal of road maintenance, improvements and investment decisions. A major
limitation of HDM4 is that it cannot be used for low-volume roads; suitable for use of roads with 200
or more vehicles. Other limitations include the model cannot capture all benefits of rural road
investments, not customised for unpaved roads and requires huge data inputs which may be
impractical to collect for rural roads (Archondo-Callao, 2004).
The World Bank developed a separate economic evaluation model, the Roads Economic Decision
(RED) model, suitable for the characteristics and needs of low volume rural roads. The model is
adapted for low-volume unpaved roads, with traffic volumes between around 50 and 300 vehicles
per day. The model performs an economic evaluation of road investment options using the
consumer surplus approach and is customised to the characteristics of low-volume roads such as the
high uncertainty of the assessment of the model inputs, particularly the traffic and condition of
unpaved roads, the importance of vehicle speeds for model validation, the need for a
comprehensive analysis of generated and induced traffic, and the need to clearly define all accrued
benefits (Archondo-Callao, 2004).
Despite its popularity and wide application, CBA has many serious limitations which are well
documented in the literature. The conversion of nonmonetary costs and benefits into monetary
values is questionable. Another major criticism is that CBA is essentially an economic efficiency
criterion; it does not consider equity or fails to capture government’s policy objectives, for example,
social and economic wellbeing of people. There are also other difficulties in applying CBA; some of
these include, obtaining reliable information on costs, estimating the costs and benefits of
externalities, estimating benefits due to shift of demand (which may be very significant especially in
case of rural roads), and accounting for the impacts outside the geographical scope of the project.
Also, the CBA methodology is not considered suitable for low-volume rural/village roads.
However, as mentioned in the previous section on MCA, CBA may also be combined with other
methodologies to overcome its limitations as well as to make the methodology more appropriate in
Page26
a given context, for example, the 2nd Phase of Rural Roads Markets Improvement and Maintenance
Project (RRMIMP-II) in Bangladesh which were implemented in 21 districts of the country (LGRDC,
2002). More details are provided in section 2.3.
2.2.3. Participatory approaches

Participatory approaches to planning increases the likelihood that actions taken or services provided
by public agencies more adequately reflect the needs of people. Public participation is understood as
a process in which all concerned stakeholders including the beneficiaries and affected people are
involved in decision-making about development works. Considering the merits of such approaches
models have been developed for planning of infrastructure facilities and various public services. In
the context of rural infrastructure development probably the most well-known is the Integrated
Rural Accessibility Planning (IRAP) methodology developed by the International Labour Organisation
(ILO).
IRAP is an area-based planning process to identify the actual access needs of the local communities
and define and prioritise interventions to improve access. Interventions to either improve rural
mobility (such as roads, bridges and tracks) or improve the distribution of facilities and services (such
as water supplies, schools and health centres). The whole process is participatory and involves the
local communities and local government officials representing the different sectors. In short, the
IRAP methodology calculates accessibly Indicators (AI) by multiplying the number of households (N)
with the subtraction of the average travel time to a facility (T) minus the acceptable/target travel
time Tm, times the frequency of travel (F): AI=N*(T-Tm)*F. Finally, maps are established with the
available information. The larger the value of the AI, the greater the problem and higher the priority
for improving access to that facility.
Further details of the model can be found in Donnges (2001) and elsewhere. The IRAP methodology
has been applied in the Philippines and many other countries for village level accessibility planning.
Sarkar and Dash (2011) developed a modified version of IRAP and applied it in India.The major
limitation of the participatory approach to planning is that it requires an institutional setting to apply
the approach. In the absence of such an institutional setting, as in the case of Bangladesh, there is
little chance of executing such approach or implement the outcome of any such exercise. These
approaches are more suitable in situations where local level planning is fully institutionalised within
local bodies.
2.2.4. Network planning models based on optimisation techniques
CBA and Prioritisation Ranking (PR) or MCA techniques may not be always suitable for the selection
of rural road projects. In both the cases there is an implicit assumption of project independence.
However, this assumption may not be valid especially in the early stage of rural road network
development. A particular road (or link) in the network may not get priority for development
considering its own merits but could be a vital link of the network. Without its development the
network as a whole may only be partially functional and therefore of limited utility to the people. In
other words, the functional utility of many roads in the network and the network as whole are
dependent on this particular road of lesser priority.
Researchers have applied some network development models such as the Minimum Spanning Tree
(MST)and other network development models (which are based on Graph Theory) to address this
issue of independence.In short, an optimum network of roads is identified by applying the MST
model which minimises the total length of the road network in connecting all villages with the
nearest market/service centre by linking the village directly with that centre, or to another village
Page27
which is already connected to such a centre, or to an existing road link which in turn serve as a link
to a centre. Good examples of network development models in rural areas of Bangladesh and Nepal
can be found in Oudheusden and L. R. Khan (1987) and Shrestha et al. (2013).
The major limitation of such models is that if the network is already developed more than the basic
minimal network to connect all market/service centres, as would be the case in most parts of
Bangladesh, such models do not work well. Another limitation is that because of their analytical
sophistication, usually it may not be easy to apply such models in, or generalise for all situations.
2.2.5. Hybrid methodologies – Combination of multiple methodologies

All of thebroad categories of methodologies discussed above have their weaknesses as well as
advantages over the other categories. In many respect, the suitability of these methodologies also
depends on the context or situation. In order to overcome their methodological weaknesses,
especially in a given context, often a combination of methodologies has been suggestedby
researchers. For example, Guhnemann et al. (2012) combined MCA and CBA to capture the best of
both. In another example, Shrestha (2003) in his methodology for district road network planning and
prioritisation in Nepal has used a network planning model, CBA, and a prioritisation index based on
four criteria (agricultural potential, interaction, accessibility and environmental impact.
Another strand of work found in the literature is the application of GIS-based models incorporating
accessibility planning. For example, Singh (2010) developed a methodology for rural road network
planning model for developing countries. The model is based on the concept of accessibility and was
applied in Paithan Taluka, India. An index of accessibility was designed which evaluated various rural
road link options for their efficiency in accessing the missing functions in the unconnected
settlements. A GIS-based technique for the analysis of alignment of new road link options was
developed which considered the topographic and land use characteristics of the area. This approach
offered maximum benefit to the unconnected settlements in terms of access to various facilities or
the main road network.
2.3. Prioritisation of rural roads in Bangladesh: Current practice

As mentioned earlier LGED is responsible for developing and maintaining rural roads in Bangladesh.
At present LGED does not have a standardized methodology for prioritisation of rural roads, though
for maintenance of road they currently follows a two-step process. It identified several indicators,
classified and assigned rating to the indicators (Table 2.1). At first, numbers of all the facilities along
the roads are identified and each indicator/group of indicators are then multiplied by their
respective assigned rating value. In the next step, summation of all the scores provides the total
score of the road. Prioritisation is determined through comparison of the total points of all candidate
roads for maintenance. Roads with higher score are given higher priority.
Guidelines for Rural Roads and Culverts Maintenance (LGRDC, 2013), provides LGED a good policy
support in maintaining its infrastructures and assets throughout the country. It provides a checklist
for considerations while prioritising maintenance requirements.Higher priority is given for periodic
maintenance, higher order roads, donor funded roads, roads having higher traffic volume and bus
routes, continuous and uninterrupted link providing roads, roads providing accessibility to more
growth centres, villages, and service/activity centres.
Page28
Table 2.1: Ratings of Indicators used by LGED for Prioritisation of Road
Indicator/Group Description Rating*
Classification Upazila Road 12
Surface Type Fully BC 0
Gaps Major Gaps 3
Traffic Volume AADT 1000+ 100
Growth Centre 12
Market
Rural Market 6
Hospitals Upazila Health Complex 9
Union Parishad Office 12

Social Centres
Other Public Centre 6
College 9
Educational Secondary School 6
Primary School/Madrasah 3
Source: RDBMS software, LGED
It is worth mentioning here, the BUET team was informed during discussion with LGED officials that
for CARE funded projects a methodology was developed for calculation of benefits of improvement
and further improvement of roads in the project area. To calculate the benefit, the methodology
only considered the types and amount of agricultural items produced within one kilometre radius
along the road. The methodology has not been used for in any other project of LGED.
An attempt was made to prioritise Feeder Road type Bs (FRB1) and important Rural Road RR1s2roads
to be improved and maintained under the 2nd Phase of Rural Roads Markets Improvement and
Maintenance Project (RRMIMP-II) which were implemented in 21 districts of Bangladesh (LGRDC,
2002). Initially the districts were ranked on the basis of spatial characteristics, demographic
characteristics, socio-economic characteristics, status of infrastructure, revenue contribution,
institutional capacity and governance and community/public participation. The roads in districts with
higher priority were considered for further analysis. The consultants identified six factors important
for prioritisation of roads. However, due to the limitations of data, only two criteria were considered
for prioritisation of important RR1s: community ranking and Benefit-Cost Ratio.
To prioritise FRBs for improvement, six factors were considered important: FRB Status (core /non-
core), growth centre connectivity (primary / secondary), growth centre development status
(developed / undeveloped) and community rank (numerical value, provided by community, of
preferences and priorities in plan implementation and maintenance programmes within their local
area), Benefit - Cost Ratio and status of connecting FRA.
1
FRB stands for Feeder Road (Type B), which has been renamed as Upazila Road (UZR) in the categorisation of
rural/LGED roads in Bangladesh
2
RR1 stands for Rural Road (Type 1), which has been renamed as Union Road (UR) in the current categorisation of
rural/LGED roads in Bangladesh
Page29
LGED conducted several monitoring and evaluation report to identify the impacts of its Rural
Infrastructure Improvement Projects (RIIPs). These reports summarise the impact of the projects
comparing baseline and end of project condition in project and control areas with respect to several
socio-economic indicators (LGED 2010; LGED 2011). These reports used several sources for inputs
used for economic analysis assuming 6% growth of transport. The Economic Analysis (EIRR) has been
carried out for 20 years of economic life, on the basis of total investment cost including construction,
land acquisition, environment mitigation measures and annual maintenance cost; and Vehicle
Operating Costs (VOC) savings with and without passenger time savings (LGED 2010; LGED 2011).
2.3.1. Merits and demerits of the current practices

It is clear that LGED uses no prioritisation methodology for improvement or further improvement of
roads. A CARE project once used a methodology which used a very simple procedure for assessing
the potential benefits of road improvement; besides it was applicable for the project area alone.
Similarly, in RRMIMP a very simple methodology was used for prioritising roads in 21 districts.
Methodology used for prioritisation of roads for maintenance has some rational basis but the scores
that are given to indicators are subjective; discussion with LGED officials reveals that these scores
were subjectively put by some experienced officials.
Prioritisation methodology used in RRMIMP study considered only the improvement of FRB and RR1
networks. In fact, while making a generalised methodology in a context where all necessary data are
not readily available throughout the country and where field context differs from one place to
another, adopting a sophisticated CBA methodology is not practical. Therefore considering a simple
methodology by RRMIMP is justified. Moreover, in RRMIMP all the FRBs and RR1s were considered.
Recognizing the ‘vast extent of work’ the scope was restricted to ‘Core’ FRBs and ‘Important’ RR1s3.
The main advantage of RRMIMP technique is that the total score of a road under consideration is an
absolute value and therefore allows independent calculation of scores. This is useful when new
roads need to be added into the priority list. However, the major shortcomings of RRMIMP
methodology are: (a) there is no rational basis for scoring the indicators, scores are given
subjectively; (b) indicators are not compatible with LGED’s current RDBMS platform; (c) no
interfacing of the database with GIS platform; (d) the methodology has been developed for a
particular project and for a particular area, application of the methodology has not been considered
for the whole country.
2.4. Rural road development programme in some Asian countries

Many countries in Asia have implemented impressive rural road development programmes. In India,
for example, a rural roads programme, the Prime Ministers Rural Roads Scheme (PMGSY), was
launched as a 100 per cent Centrally Sponsored Scheme, with the objective of boosting rural
connectivity. The aim of the scheme was to provide habitations with a population of 1,000 and
above, as well as those with a population of 500 and above, with all-weather roads. In
hilly/desert/tribal areas, its aim was tolink habitations with populations of 250 and above. In 2002,
there were a total of 178,768 habitations falling within these categories. During the Tenth Five Year
3
Core FRBs has been defined as those which satisfy the functional definition of FRB namely; roads which connect
Growth Centre markets at one end with Upazila Headquarters or RHD road network or another Growth Centre or
another FRB at the other end, while all RR1s identified by the Upazila Engineers have been considered as
important RR1s.
Page30
Plan (2002-2007), 88% of the targeted habitations were connected (GOI, 2008). A manual prepared
by the National Rural Roads Development Agency (NRRDA, 2005) provides the operational details of
the PMGSY programme.
The programme has several innovative features which include: first that it is financedfrom a small
dedicated levy on diesel fuel; second, that the engineering design and construction standards are set
by the central Government; and third, that physical and financial implementation of the scheme is
done through a centralised computer system. In an evaluation undertaken by the Planning
Commission of India it was observed that PMGSY had succeeded in providing connectivity to some of
the most deserving habitations (Government of India, 2008). In 2007, PMGSY was folded into the
Bharat Nirman programme. This is a more broad-based programme that includes all rural
infrastructure development including rural roads. With respect to rural roads, it aimed to connect all
1,000-plus habitations in rural areas (500-plus in hilly and tribalareas) by 2012.
The Gama Neguma programme in Sri Lanka is designed to achieve island wide economic
development through the creation of economically prosperousvillages. The programme was
launched in 2006 in 119 Grama Niladhari divisions and will ultimately cover all of the 14,034
divisions. Under the programme, 3,354 km ofroads and bridges were built in 2008 at a cost of 7,058
million Sri Lanka rupees (Government of Sri Lanka, 2008).
Bhutan has constructed priority feeder roads in five districts, where isolated rural communities had
to walk for one to three days to reach all-weather roads or markets, schools, health clinics, hospitals
and district headquarters (IDA, 2008).
The Philippines is implementing an Intensified Building-Up of Infrastructure and Logistics for

Development (I- BUILD) programme to develop a network of climate-resilient rural infrastructure
and facilities which include farm to market roads and bridges and other infrastructures (Philippines
Rural Development Project, 2015). I-BUILD is centrally managed and funded with the support of the
World Bank, and is implemented in direct collaboration with the Local Government Units (LGUs). The
proponent LGUs, among other things, must be willing and capable to contribute to funding
requirements and have the technical capability to plan and implement projects.
The actual project selection and implementation is done by the LGUs. There are a few general and
some specific criteria for project selection. However, as funds available may not be sufficient to
implement all such selected projects, a set of specific prioritisation criteria is applied for the ranking
and final selection of projects. These prioritisation criteria include:
 Number of beneficiaries – the bigger the number, higher the priority.

 EIRR – at least 15%; the higher the EIRR, the higher the priority
 Benefit-Cost Ratio – at least 1.0; the higher the B/C ratio, the higher the priority
 Per capita subproject cost – the lower the cost per beneficiary, the higher the priority
 Subproject location – higher priority shall be given to subprojects located in LGUs that had
no or limited government assistance similar to the type of subprojects being proposed
 Gender and/or children sensitivity – all other things being equal, a subproject with more
women and/or children to be benefited shall have higher priority
 Level of poverty – if reliable statistics are available, the higher the poverty level, the higher
the priority
 Projects with established right-of-way (ROW) have higher priority.
Page31
The programme has established a mechanism to ensure the operation and maintenance (O&M) of
the implemented projects by the LGUs. O&M audit results are used to verify compliance. Sanctions
apply for non-compliance of operations and maintenance requirements.
In China, about 2 million kilometres of roads classified as rural were constructed during the period
1996-2004. A feature of the interventions in China is that they provide not only transport
infrastructure but also transport services. The road development strategy envisages that regular bus
services will serve no less than 95 percent of villages by 2010 and all villages by 2020.
Many countries have implemented impressive rural road development programmes. However, often
they have not been equally successful in establishing a sustainable maintenance programme, which
is vital not only to preserve the value of national assets, but essential to provide continuing support
to the growth of rural economy. In the Asia-pacific region, China has a high rate of rural road
maintenance, with an estimated 90 per cent of rural roads are currently maintained. It has
established an institutional setup involving financial and implementation arrangements and
responsibilities at different tiers of the government for proper maintenance of rural roads (Hongye
Fan, 2012).
2.5. Observations and conclusions related to developing a methodology for the

present study
Several conclusions can be drawn from the discussions presented in the preceding sections that can
be useful in developing a planning and prioritisation methodology for rural road development in
Bangladesh.
The national programmes of road development generally are based on some broad criteria that can
be easily applied all over the country to develop a rural road network. The main purpose of
developing such a network is to provide access to all habitations of a sizable population by an all-
weather road. However, given the usual budget constraints, the actual selection of projects at the
implementation level may be based on merits of the competing projects. Such merits of individual
projects may be considered by the outcome of some form of economic or other types of evaluation.
It is quite common to see adoption of some simple type of evaluation tool such as priority ranking
based on a few criteria to prioritise road projects. However, CBA and Prioritisation Ranking (PR)
techniques alone may not be always suitable for the selection of rural road projects.
On the other hand, many network models based on some optimisation techniques can overcome
limitations of CBA or PR based approaches. These models also have intellectual appeals. However,
the major limitation of such models is that if the network is already developed more than the basic
minimal network to connect all markets/service centres, as would be the case in most parts of
Bangladesh, such models also do not work well. Another limitation is that because of their analytical
sophistication, it is not easy to apply such models in all situations and cannot be generalised. Their
application generally requires substantial amount of data collection and the models can be
developed generally for a particular given area and in a given context. Consequently, their
applications have remained limited mainly to research studies.
Often participatory approach to planning has been touted to be a preferred approach especially for
local level planning such as rural infrastructure development. However, the major limitation of the
participatory approach to planning is that it requires an institutional setting to apply the approach. It
may work well where local level planning has a strong footing such as in the Philippines. In the
Page32
absence of such an institutional setting, as is the case in Bangladesh, there is little chance of
implementing this approach or execute the outcome of any such exercise. These approaches are
more suitable in situations where local level planning is fully institutionalised within local bodies.
The application of Cost Benefit Analysis (CBA) has been widely adopted in analysing the merits of
road investment projects in both developing and developed countries. CBA has been popular among
the decision makers in Bangladesh as it provides an easily understood framework to evaluate all
costs and benefits of a project, and CBA’s appealing evaluation criteria such as EIRR or B/C ratio
values. For being a preferred choice among the decision makers in government, LGED has requested
the project team to develop a simplified CBA based on limited data for the economic evaluation of
rural roads. It is noted that the development of a simplified CBA methodology for the evaluation of
rural roads is a strategy stated in the Seventh Five Year Plan of Bangladesh (see section 1.1).
Despite its popularity and wide application, CBA has many serious limitations which are well
documented in the literature. The conversion of non-monetary costs and benefits into monetary
values is questionable. Another major criticism is that CBA is essentially an economic efficiency
criterion; it does not consider equity or fails to capture government’s policy objectives, for example,
social and economic wellbeing of people. In order to overcome such limitations, CBA analyses
should be supplemented by other types of analyses such MCA that can address some the limitations
of CBA, especially in case low-volume rural roads.
Multi Criteria Analysis (MCA) or Prioritisation ranking (PR) techniques of rural roads have been
widely used for the selection of rural road projects in many countries of Asia and Africa. A set of
factors or criteria reflecting policy or social objectives of the government is selected for ranking of
projects.The MCA methodology is used to rank rural road investments, and is typically applied when
traffic volumes are too low to apply the conventional cost benefit analysis to make sense, but there
is a strong belief that there will be important social benefits arising mainly from improvement of
access and connectivity. Many rural roads in Bangladesh are low-volume roads and mainly used by
pedestrians. Typical CBA analysis for such roads will not produce any meaningful results that may be
used for investment decisions. The application of some form of MCA would be more appropriate in
such a situation.
Besides choice of planning and investment decision-making tools as discussed above, there are other
considerations that may also need to be considered while developing a suitable methodology.
Bangladesh already enjoys a high density of rural roads compared to other countries in the region.
However, most of the roads are not all- weather roads and many villages are not connected by all-
weather roads.
LGED has already developed a network of about 74,000 km of paved roads and about 20,000 km of
brick paved roads which may be paved in near future. Any further development of roads should be
selective and based on some criteria so that the network can be sustained and properly maintained
over a longer term. The selection criteria, among others, may include considerations of access to all
villages and connectivity with higher order centres. Some broad criteria, similar to national
programmes in other countries, can be considered to develop core networks of rural roads at
Upazila and Union levels which can provide rural accessibility at all levels as well as ensure intra- and
inter-upazila connectivity. Fortunately, such national level broad policies are already in place that
may be considered to develop such core networks. These are discussed in the following chapter on
methodology.
Page33
3 Methodology for the Study
3.1. Introduction
The Project team developed a methodology for the present study and its future application by LGED
in other districts of Bangladesh. The details of the methodology, as applied in this study, are
discussed in this chapter.
In developing the methodology the following matters were considered:

 Government objectives, current policies and strategies as stated in official documents
 Guidance and advice of the Working Group established for the project
 Comments and suggestions received at the First Stakeholders Workshop
 Consultation meetings with concerned LGED officials at the HQ and field offices, and current
practices of LGED
 Observations and conclusions distilled from the literature review
The project team had many rounds of consultation meetings with LGED officials in Dhaka, and met
field-level government officials, elected representatives, local leaders and other knowledgeable
persons in two upazilas of Tangail, the selected pilot district. The main purpose of such consultation
meetings were:
 To have a clear understanding of the current prioritisation practices at LGED and issues in
rural road development;
 Requirements of LGED;
 Opinion of local leaders, people and other stakeholders on benefits and impacts of rural
roads and their prioritisation; and
 Availability of information from secondary and other local sources.
The project team also considered the following important government policies, conditions and
assumptions in developing the methodology:
 Government policy on rural roads as outlined in the Seventh Five Year Plan (SFYP) (see pp.
392-93) and other official documents;
 It is envisaged that generally no new rural roads will be constructed. The existing roads, as
needed and prioritised, may be improved, further improved (or upgraded) and maintained;
 The current prioritisation practices of LGED, methodologies applied in previous studies such
as the Regional Road Network Study for prioritisation under RRMIMP-II (LGED, 2002), and
Guidelines of the Project Appraisal Framework (PAF) prepared by the Planning Commission
(Planning Commission, 2005);
 The selection of rural roads for development should be based on a set of prioritisation
criteria which should reflect road investment costs, usual direct benefits to road users
(where such estimation is possible), improved accessibility to markets and social
infrastructure facilities, connectivity, pass ability of the road in all seasons, and other local
priority of special nature (for example, access to cyclone shelters, access to “ghats” and
public transport stops, stations etc.);
 Minimum new data collection, which should be available from secondary sources such as
LGED and other government departments and other local government sources. Unless
otherwise essential, primary data collection should be avoided.
Page34
3.2. Road categories and types of road development
An upazila may have different categories of roads – national and regional highways and zila roads
that are managed by the Roads and Highways Department (RHD); and upazila, union and village
roads managed by LGED. For the purpose of the present study, roads managed by RHD have been
considered as given roads and only the roads managed by LGED have beenconsidered for
development.An example of an upazila road network with different types of roads and their status is
shown in figure 3.1.
Figure 3.1: Upazila road network with different types of roads in Mirzapur, Tangail District
Source: LGED GIS Database, 2010
Following initial discussion with LGED officials and later on at a Working Group meeting it was
decided that methodologies on prioritisation would be developed for three types of road
development works namely, improvement, further improvement/upgradation and maintenance of
rural roads. The definitions of these three terms are as follows:
Improvement
 Converting an earthen road to a road i.e., from earthen to BC/RCC in an existing alignment
 Converting a partly paved road to fully paved road
 Converting an HBB road to fully paved road
 Improvement of road geometric standards, raising of embankment and widening of
pavement and/or crest and raising of road embankments of an existing road
Maintenance
 Maintenance of an already paved road (BC, RC or HBB).
Page35
3.3. Major components of the methodology for the study
The methodology has two major components. The first component follows a network approach to
road planning; and the second component involves prioritisation of road development based on a
set of physical, economic, social and other criteria (Figure 3.2). Both the components include the
provision of local stakeholders’ participation in network planning and prioritisation.
All the designated Upazila and Union roads along with the national roads, if any, form the core
network in an Upazila. The core network also includes roads that crosses the geographical boundary
of an Upazila to connect an activity centre or a national road in a neighbouring Upazila. In the
prioritisation scheme all such roads have been given additional importance as together they form a
network vital to maintain inter- and intra-Upazila transport connectivity. The activity centres form
the nodes and Upazila and Union roads connecting them are the links of the core network.
Figure 3.2: Schematic diagram showing overall approach to planning and prioritisation of rural roads
Planning and Prioritisation of Rural Roads
Review of current policies, practices and resources
Updating of Databases
LGED Databases
 Road Database Additional Data
Collection
 GIS Database
Local workshop of
stakeholders
Connectivity, local
priority, road Multi-Criteria and
safety Cost-Benefit
Analyses
Component 1 Component 2
Network Planning- Prioritisation of
Core Network rural road
development
Page36
The core road network has two components: upazila-level core network, and union-level core
network. The upazila core road network serves upazila-wise access needs and ensures connectivity
with the neighbouring greater region/rest of the country. The union core network serves union-wise
and inter-village access needs and ensure connectivity with other areas of the same and
neighbouring upazila(s).
The prioritisation component of the methodology considersphysical (including connectivity),

economic, social and some other aspects with their due significance. A set of prioritisation criteria
for the assessment of costs, benefits, and social and environmental considerations been selected.
The selected criteria include road construction costs, direct benefits to road users (where such
estimation is possible), improved accessibility to markets and social infrastructure facilities,
connectivity, passability of the road in all seasons, and other local priority of special nature.
Further details on these two components of the methodology are presented below.
3.4. Network planning - Defining the core upazila and union road networks and
identification of priority road development needs
The preliminary upazila and union core road networks may be identified from the information
available from LGED’s databases. The roads of the upazila core network connect an upazila with
other upazilas and important activity centres within the upazila. The RHD Roads (if any) and Upazila
Roads are vital for ensuring connectivity of an upazila with the greater region/rest of the country.
Union roads are important as they connect Union HQs with Upazila HQs, Growth Centres or local
rural markets. Therefore, all RHD Roads, Upazila Roads and Union Roads will be included in the core
network. However, some village roads providing access to a large number of socio economic
facilities and rural markets are also important for addressing the union-wise access needs. In
summary, the preliminary core network comprises of:
 All RHD Roads (the given roads)

 All upazila roads that directly connect to national and regional highways of RHD, zila and
upazila headquarters and other higher order national centres, and railway and steamer
stations
 All upazila roads that directly connect to designated growth centres in an upazila or a
neighbouring upazila, zila roads of RHD, and that crosses the geographical boundary of an
upazila to connect an important centre or a national highway or zila road in a neighbouring
upazila
 All union roads that directly connects to upazila and union headquarters, important market
centres in the same or neighbouring upazila, and zila roads.
 Other Union Roads
 Some village roads having the facility score above 50 on a scale of 100.
Facility scores are calculated in terms of total number of socio-economic facilities (schools, hospitals,
industries and other important centres), growth centres (GCs) and rural markets (RMs).
Page37
3.4.1. Validation of the preliminary upazila and union core road networks
Local workshops attended by the local leaders, officials and other stakeholders were organised one
in each of the 12 upazilas of the pilot district. Further details about the organisation of local
workshops are provided in Chapter 5. The local workshops, among others, had two major objectives:
 Validation of the preliminary upazila and union core road networks, and
 Identification of roads according to their level of local priority for development
The workshop participants scrutinised the preliminary upazila and union core road networks
presented to them. This was done to ensure that no important road that should be part of the core
networks was left out. For example, roads that connects important activity centre
(GC/RM/Ghat/Station or any other important economic/agricultural/ industrial or other
establishment) or links of the already identified core network with higher order roads were included
in the core network. Some village roads having many facilities by them (with a facility score of 50 or
above) were included in the core network. In this process, the preliminarily defined core networks
were finalised at the local workshops.The activity centres (Upazila and Union headquarter, growth
centres, rural markets) form the nodes, and upazila and union roads connecting them are the links of
the core network. Reports pertaining to the local workshops are provided in Annex F, volume 2.
3.4.2. Connectivity and local priority for development

Connectivity of roads was identified as an important criterion for their prioritisation. Accordingly,
connectivity was included as a criterion for road prioritisation evaluation by MCA (discussed later in
the section; see Tables 3.2-3.3). Roads in the core network which provide higher order connectivity
get a higher score for priority. Upazila roads that directly connect to national and regional highways
of RHD, zila and upazila headquarters provide the highest order connectivity because they connect
the entire upazila with the greater region and rest of the country. Other Upazila Roads that connects
one growth centre to another growth centre, Union Roads and the village roads selected as part of
core networks get the priority scores according to their respective order of connectivity as shown in
Table 3.1.
Table 3.1: Level of connectivity of roads and their assigned scores

Level of connectivity Road Score
Part of core networks Upazila roads that directly connect to national 100
and regional highways of RHD, zila and upazila
headquarters
All other Upazila Roads that connects one 80
growth centres to another growth centres.
All Union Roads 50
Village roads those are part of core network 35
Not part of core networks Village roads those are not part of core 0
network
The workshop participants also identified the roads according to their level of priority for
development. This is termed as ‘local priority’ which is one of the criteria for MCA for road
prioritisation. In addition, the local stakeholders also identified some roads from among the upazila
and union roads which may need further improvement/upgradation in terms of geometric standards
Page38
such as road width and/or pavement strengthening on a priority basis. The stakeholders also
discussed and suggested what type of “improvement” may be needed for such prioritised roads.
Although collection of these suggestions was not necessary for the present study, this was
undertaken with a view to assist the local LGED offices. The Upazila Engineer can consider these
suggestions while undertaking detail design for such prioritised roads when funding becomes
available.
3.5. Prioritisation of rural roads development
The prioritisation methodologies on improvement, further improvement/upgradation and

maintenance of rural roads covers all three categories of rural roads namely, Upazila Roads (UZR),
Union Roads (UNR) and Village Roads (VR). The type of development depends on road type, surface
condition, roughness, and traffic volume. Road development by category of roads are as follows:
 Unpaved roads require improvement meaning converting them into paved road
 Partly paved and HBB roads require both improvement and maintenance
 Fully paved roads require maintenance.
 In addition, some fully paved roads with high traffic volume may require further
improvement or upgradation in terms of geometric standards and/or pavement strength.
The methodologyconsidersthe evaluation outcomes of both the Cost Benefit Analysis (CBA) and
Multi-criteria Analysis (MCA) for prioritisation of roads. Prioritisation is done by applying MCA using
a selected set ofsocio-economic criteria. The selected criteria include traffic volume, access to
markets and social infrastructure facilities, connectivity, local priority of special nature and a few
other factors. The selection of criteria for MCA, however, depends on the type development and
category of road. Separate sets of criteria are used for each type of development.
In addition to MCA, Cost Benefit Analysis (CBA) has been considered to evaluate the economic
merits of road development. However, CBA is not applied for prioritisation of road maintenance. An
already paved road requires maintenance anyway. Only MCA has been considered for prioritisation
of maintenance.
The flow diagram in Figure 3.3 shows how a road is selected (from among roads in an upazila) for a
particular type of road development. There are specific criteria for the selection of a road for each
category of development. These criteria have been chosen so that they can be used in a software to
select all roads that may qualify for prioritisation analysis under the four broad groups as mentioned
below:
1) All unpaved roads for improvement (converting into paved roads)
a. High traffic volume
b. Low traffic volume
2) Partly paved roads
a. Improvement (paving the unpaved part)
b. Maintenance (IRI value or roughness condition assessed by visual inspection)
3) Fully HBB
a. Improvement (converting into BC/CC/RCC)
b. Maintenance (roughness condition assessedby visual inspection)
4) Fully paved roads
a. Further improvement (for roads with peak hour PCU > 290 or CVD > 300 VPD)
Page39
b. Maintenance (IRI value or roughness conditionassessed by visual inspection)
Prioritisation analysis is undertaken for each of the above 8 sub-categories of roads by applying CBA
and/or MCA. Further details on these analyses are provided hereafter.
3.5.1. Prioritisation for “improvement” of rural roads

All unpaved or earthen roads are considered for development under the category “improvement”.
Besides, the partly paved roads that needed to be fully paved are included in this category. Also
included are the HBB roads that needed to be upgraded into BC or RCC. The local situation would
determine the appropriate type of improvement. For example, in “haor” and some coastal areas,
where roads may remain submerged for a considerable time, RCC may be right type of pavement.
For unpaved roads, low volume roads need to be considered separately from high volume roads for
prioritisation. The majority of the traffic of low volume rural roads consist of pedestrians and Non-
Motorised Traffic. The benefit due to lower transportation cost resulting from the development of
the roads would be small. Therefore, these roads are associated with less economic merit than
others.
Annual Average Daily Traffic (AADT) value is proposed to cut-off the low volume roads from high
volume ones. The cut-off value of AADT was determined after CBA of rural roads with different
traffic volume. It showed that low volume unpaved roads having AADT value less than 200 has less
economic merit. Therefore, an AADT value of 200 was considered as a cut-off value. It may be noted
here that the RED software developed by the World Bank also suggests such an approach.
Multi-Criteria Analysis (MCA) in conjunction with Cost Benefit Analysis (CBA) were used for the
prioritisation. Different socio-economic criteria were considered in MCA. Evaluation criteria used in
the MCA for road improvement are listed out in section 3.6 (Tables 3.2 and 3.3). As required by
LGED, Economic Internal Rate of Return (EIRR) values were calculated for all unpaved roads. The
details on the CBA are provided in Section 3.7.
The MCA score suggests the importance of a road from socio-economic considerations. On the other
hand, CBA results (EIRR value) may be used mainly to determine which roads should get highest
priority for improvement considering their economic merits. The two evaluation results (MCA score
and EIRR value) can be standardised to combine the results, which may be used to prepare a
combined priority ranking of roads for the improvement of different types of roads. It is important to
note that the relative importance of the socio-economic context (MCA) and economic merit of the
project (CBA) may vary depending on the context of development and availability of funding. The
ultimate decision rests with the decision makers, however.
Page40
Figure 3.3: Flow chart showing selection roads from the road database for their evaluation
Page41
3.5.2. Prioritisation for “further improvement”/Upgradation of rural roads
Roads which are already paved but requires upgrading in terms of geometric standards (widening,
for example) or pavement strengthening due to changes in traffic volume and/or their composition
(considerable increase in number of heavy vehicle, for example)are considered under this category
of development. The methodology considers paved roads with more than peak hour PCU of 290 or
300 CVD for further improvement. It may be noted here that these values are also consistent with
the LGED’s current road design standards.
Multi-Criteria Analysis (MCA) in conjunction with Cost Benefit Analysis (CBA) has been used for the
prioritisation of further improvement of roads.The MCA for further improvement (upgradation) are
based on socio-economic and other criteria (similar to earlier described MCAs for road
improvement) along with other criteria which include traffic volume (AADT), road geometry, surface
type (BC or RCC) and road category (UZR, UNR or VR). The prioritisation criteria used in MCA for
further improvement of roads are provided in Table 3.4.
3.5.3. Prioritisation for “maintenance” of rural roads

The roughness condition of a road is considered to determine if it needs maintenance. The
International Roughness Index (IRI) value, if available,has been usedto determine the roughness
condition of paved roads (RCC and BC roads) and partly paved roads (where maintenance of only the
paved segments are necessary). If IRI values are not available, the qualitative assessment
information available in the road database based on visual inspection of road surface conditions
have been used. The qualitative assessment is based on the number of potholes, cracks, depression
and edge failure, and ravelled and rutting conditions observed during visual inspection. The
existence of these features are collected and are included in the road database.
The prioritisation of road maintenance is based on the outcome of an MCA. The MCA considers a set
of criteria which include traffic volume (AADT), last maintenance year, road surface type, as well as
some soci-economic and other factors. The choice of these criteria follows the current practices of
LGED and the Guidelines for Rural Roads and Culverts Maintenance prepared by the Ministry of
Local Government, Rural Development and Cooperatives (LGRDC, 2013). Based on the MCA scores,
priority lists are generated for Upazila, Union and Village roads. The prioritisation criteria considered
in MCA for road maintenance are given in Table 3.5.
3.6. Multi-Criteria Analysis
The Multi-Criteria Analysis (MCA) forms an important part of the prioritisation methodology. An
MCA or an MCA in conjunction with a Cost Benefit Analysis (CBA) have been used for prioritisation of
roads. Initially a set of criteria, reflecting policy and social objectives of rural road development,
wasidentified considering LGED’s current practice (see section 2.3 in literature review), views of
experts and literature review. A survey was conducted among the stakeholders (including LGED
officials) to identify the potential criteria from among this initial list (see, first stakeholders meeting
report). After further discussion with LGED officials and other experts, these potential criteria were
considered by the Working Group at a meeting and were finalised. The selected criteria include
variables such as access to markets and services, connectivity, traffic volume, social and economic
factors, etc.
After selection of the MCA criteria, their relative weights were established. The Analytical Hierarchy
Process (AHP) technique, which is commonly used for the establishment of relative weights of
factors, was used for this purpose. A survey instrument was designed and a survey was conducted
Page42
among the experts and LGED officials at the HQ and field offices to determine the relative weights of
the selected criteria for their use in MCAs. The survey instrument is shown in Annex G, volume 2.
Finally, the priority score or index value of each road was calculated by summing the product of the
actual value of each criterion (the actual value was standardised) by its relative weight. This section
outlines the evaluation criteria that were considered for undertaking the MCAs for each type of road
development.
3.6.1. Selection of evaluation criteria for Multi-Criteria Analysis (MCA)
MCA for improvement of Unpaved Roads
The benefits of improving Upazila, Union and Village roads may be estimated in terms of savings in
vehicle operating costs and travel time cost. A CBA methodology has been applied to estimate such
benefits. However, these roads also provide access benefits to the people living along such roads
many of whom may not be using any vehicular traffic. In order to make a more complete
assessment, the access benefits to such people also need to be assessed by using an MCA. For MCA
of unpaved roads, traffic volume, number of socio-economic facilities, growth centre and rural
market served by the road, connectivity and local priority have been considered. Table 3.2 shows the
list of criteria used to undertake the MCA for the improvement of unpaved roads.
MCA for improvement of partially paved and HBB to fully paved roads
Partially paved roads (which have an unpaved segment) and HBB roads have also been considered
for improvement. In addition to the criteria for the MCA of improvement of unpaved roads, some
additional criteria, such as surface type, road type, and road safety have been considered for the
evaluation of such roads. Table 3.3 shows the list of criteria used for prioritisation of improvement of
partially paved and HBB roads.
MCA for further improvement/Upgradation of roads
For the evaluation of identified roads for further improvement, a set of criteria related to road
attributes such as surface type, road type, road safety plus the above-discussed criteria for the
improvement of unpaved roads have been considered for further improvement/upgradation of
roads. Table 3.4 shows the list of criteria used for prioritisation of further improvement of roads.
MCA for maintenance of roads
Roads requiring maintenance have been evaluated considering traffic volume, last maintenance
year, surface type, road type, facilities served, Growth Centres (GC)/Rural Markets (RM) served,
connectivity to higher roads and other centres will be considered. In addition, the roads serving bus
routes were given additional priority. Table 3.5 shows list of criteria used for prioritisation of roads
for maintenance.
Page43
Table 3.2: List of selected criteria for improvement of unpaved roads
Criteria Description Unit of measurement
1. Traffic Volume  Annual Average Daily Traffic (AADT) AADT (MT)

will be considered
2. Facilities served  Education Institutes, Health facilities, Number of Facilities
Industries, other public centres will be
considered as the important facilities
3. Growth centres  Hats and Bazars (rural markets) served Number of GC/RM
(GC)/Rural Markets (RM) by the roads
served
4. Connectivity to Higher  Upazila level connectivity Level of connectivity
Roads and other centres  Union level connectivity identified in the core
network
5. Local priority  Priority given by local representatives  Very high priority
 Medium priority
 Less priority
Table 3.3: List of criteria for improvement of partly paved and HBB roads
1. Traffic volume  Annual Average of Daily Commercial AADT (MT)

Traffic will be considered
2. Surface type  BC + HBB + Other fully paved Percentage of paved segment
 Fully HBB/Other paved
 BC + HBB + Other + Earthen
3. Road type  UZR will be given the highest priority, UZR>UNR>VR
followed by UNR and then village
roads (VR)
4. Road Safety  Whether should be considered for Yes/No
road safety
Industries, Other public centres will
be considered as the important
facilities
6. Growth centres  Hats and Bazars are termed as GC Number of GC/RM
(GC)/Rural Markets and RM
(RM) served
7. Connectivity to  Upazila level connectivity Level of connectivity identified
Higher Roads and  Union level connectivity in the core network
other centres
8. Local priority  Priority given by local  Very high priority
representatives  Medium priority
 Less priority
Page44
Table 3.4: List of criteria for further improvement/upgradation of roads
1. Traffic volume  Annual Average of Daily Commercial Traffic AADT

will be considered
followed by UNR and then village roads (VR)
3. Road Safety  Whether should be considered for road safety Yes/No
Industries, Other public centres will be
5. Growth centres  Hats and Bazars are termed as GC and RM Number of GC/RM
(GC)/Rural Markets
(RM) served
6. Connectivity to  Upazila level connectivity Level of connectivity
Higher Roads and  Union level connectivity identified in the core
other centres network
7. Local priority  Priority given by local representatives  Very high priority
 Medium priority
 Less priority
3.6.1. MCA scoring and weighting framework
Weights of the selected criteria
Determining the relative weights of the selected criteria is important for MCA since all criteria do not
bear the same level of importance. The more the importance of a criterion, the higher is its weight.
As already mentioned, the AHP technique has been used to determine the relative weights of a
selected set of criteria. The criteria were evaluated by pairwise comparison method. Selected LGED
officials, experts and other stakeholders actively participated in the AHP survey. The consistency of
the replies was checked by considering the consistency index (CI) value. The survey responses from
an individual that met the acceptable level of consistency (less than 20%) were used to calculate the
final weights. A spread sheet based AHP model developed by K.D. Goepel (http//bpmsg.com) was
used for the present study. Tables 3.6 through 3.9 show the relative weights of the selected criteria
used for road prioritisation.
Page45
Table 3.5: List of criteria for maintenance of roads
1. Traffic volume  Annual Average of Daily Commercial AADT (MT)

Traffic will be considered
2. Last maintenance  Roads with long gaps of maintenance Number of years between
year will be prioritised present and the last
maintenance year
3. Surface type  Current pavement type will determine Percentage of paved

nature of improvement (HBB to BC or segment
increasing width)

followed by UNR and then village roads
(VR).
Industries, Other public centres will be
6. Growth centres  Hats and Bazars are termed as GC and Number of GC/RM
(GC)/Rural RM
Markets (RM)
served
7. Connectivity to  Upazila level connectivity Level of connectivity

Higher Roads and  Union level connectivity identified in the core
other centres network
Table 3.6: Selected criteria for improvement of unpaved roads and their weight
Selected criteria AHP weight in %
Traffic volume 7.95

Facilities along a road 18.65
Growth Centre/Rural Market served 29.20
Connectivity with higher order roads 35.80
Local priority 8.40
Source: Results of AHP application by the project team
Page46
Table 3.7: Selected criteria for improvement of partly paved and HBB roads and their weight
Traffic volume 20.01
Surface type 5.70
Road type 11.83
Road safety 6.69
Local priority 5.97
Table 3.8: Selected criteria for further improvement of roads and their weight
Road type 7.86
Road safety 7.84
Local priority 14.93
Table 3.9: Selected criteria for road maintenance and their weight

Last maintenance year 4.30
Surface type 7.80
Road type 14.15
Bus Route 11.50
Weights of the Facilities
Fifteen different types of facilities have been considered in MCAs. Similar to individual criterion, all
facilities do not have the same level of importance, some are of higher order than others. The order
of importance of a facility depends on its level of influence. For the present study, the level of
Page47
influence of a particular facility was determined by the Reed-Muench ‘Median Threshold Population’
method (Haggett and Gunawardena, 1964). The higher threshold population of a facility means a
higher level of importance and therefore a higher value was assigned as the weight of that particular
facility. For example, the median threshold populations of a primary school or a community clinic
shows that their influence area is limited to only a particular village, while the influence areas of a
college or a health centre are much larger and they serve a much larger population. Accordingly,
facilities such as college, health care centre and other facilities serving a much larger population
were given higher weights compared to primary school or other facilities serving much less
population. Table 3.10 shows the weights of different types of facilities considered in MCAs.
Methodology followed to estimate population threshold is provided in Annex H, volume 2.
The score of a road for facilities served was found by summing the products of number of facilities
along the road by the corresponding weights of the facilities as shown in Table 3.10.
Table 3.10 Assigned weights of socioeconomic facilities

Types of facilities Weight
Educational Facilities Primary School 1

Secondary school 3
Madrasa 6
College 9
Health Facilities Community Clinic/ FWC 1
Union Health Center 6
Upazila Health Center 9
Non Govt. Hospital 9
Private Clinic 9
Other facilities Small Industry 1
Medium industry 3
Large industry 6
Cyclone shelter 1
Cyclone shelter cum school 1
Market Centres Other Public centre 3
Growth centre 6
Rural market 1
Source: Weights determined by using the Reed-Muench Median Threshold Population method
3.6.2. MCA scoring
The actual values of the selected criteria are in different units of measurement, and the range of
their minimum and maximum values widely vary. In order to allow for their addition, the actual
value of each criterionwas standardised on a scale of 0 to 100. The maximum value was considered
as 100 and the minimum as 0. The standardised values of actual intermediate values were
determined proportionately with respect to the maximum and minimum values. A combined score
was calculated for all candidate roads which ultimately determined the priority ranking of the road.
Page48
The combined score was calculated both in terms of the weights and standardised score of the
selected criteria. The following equation was used to calculate the combined score:
I   ( wi  xi )
I= Combined score, wi= Relative weight of criteria i, xi= Standardised score of criteria i
The priority order for the development of individual roads was determined on the basis of their
combined scores.
3.7. Cost Benefit analysis
The results of Cost Benefit Analysis (CBA) can be an important consideration for making a road
project investment decision. The present study has developed a CBA methodology for its application
in rural road prioritisation. The CBA methodology follows the VOCS and TTCS approach. It considers
the benefits of road improvement by estimating savings in vehicle operating costs and travel time
cost against the costs of road improvement and regular and periodic maintenance. It uses a standard
analytical framework of CBA to calculate economic internal rate of return (EIRR) for different types
of road improvement as discussed earlier.
Since the road users are the immediate beneficiaries of an improved road, it was necessary to obtain
information on the prevailing traffic and its characteristics to estimate road user benefits. Data on
the following items were used in CBA to estimate user benefits.
 Annual Average Daily Traffic (AADT)

 Composition of traffic
 Travel purpose and passenger occupancy in vehicles
 Vehicle operating speeds and costs by road roughness condition
 Value of time of passengers
Data on the first two items were available from the road database. Information on the remaining
three items were used from a most recent study undertaken by GTZ for LGED.
3.7.1. Projection of future traffic volume
Traffic volume is an important variable in the CBA analysis. As such, the projection of future traffic
volume is an important task. The current AADT and CVD values by type of vehicle were available
from the LGED’s road database. The volume of traffic on“hat” (market) and “non- hat” days varies
significantly. For use in CBA, the weighted average of AADTson hat (assuming two days in a week)
and non-hat days (5 days in a week) was considered.
The growth factor method wasused to project future traffic volumes of roads. The growth factor was
estimated using trend analysis of traffic growth on rural roads in recent years. It is likely that the
growth rates may vary by district/locality depending on type of economic development in the
surrounding areas. Traffic growth rates in different situations were examined and the generalised
rates of traffic growth were estimated. For the present study, 6% growth for normal traffic and
Page49
another 6% growth for generated traffic were used for traffic projection. However, it is unlikely that
such high traffic growth rates (total 12% a year) would be sustained over a long period of time.
Beyond 10 years, it is unlikely that there would be any significant generated traffic growth as most
new developments within the service area of a road would have taken place by then. Therefore, for
the purpose of this study, the growth of generated traffic after 10 years is taken to be nil. However,
it was assumed that normal traffic would continue to grow at 6% per year over the whole economic
life of a road.
As in other studies, 20 years of economic life for a road has been used. The traffic growth factor is
one of the configurable variables in the software, the values of which can be changed in future if the
traffic growth rates change. Further discussion on this matter is presented in chapter 6.
3.7.2. Estimation of costs and benefits
The cost components in the CBA analysis includes updated standard construction and maintenance
costs for different categories of roads4 and road structures as currently being used by LGED in
Tangail. Appropriate shadow factors for labour and material costs and current tax rates were applied
to calculate the economic costs of road construction and maintenance. Similar to traffic growth
factor, the values of cost variables are also configurable in the software. They can be changed for
other districts, if needed, and in the future when the costs may change.
The benefits of road improvement are estimated from lower vehicle operating costs (Vehicle
Operating Cost Savings; VOCs) and savings in travel time(Travel Time Cost Saving; TTCS). The Vehicle
Operating Cost (VOC) varies by type of vehicle and according to the roughness condition of a road.
Therefore, the amount of such benefits for each type of vehicle have been estimated considering
roughness conditions “before” and “after” road improvement.
Table 3.11: Standard parameter values used in CBA
Parameter Value
Construction cost per km (12ft/3.75m pavement width), Taka 8,000,000

Routine maintenance cost per km (12ft/3.75m pavement width), Taka 500,000
Periodic maintenance Cost per km (12ft/3.75m pavement width), Taka 1,500,000
Annual cost increment 12%
Salvage value (% of original construction cost) 25%
Tax rate 11.50%

Economic life (years) 20
Shadow price Weighted average SCF for construction cost (labour
factor (to and material) 0.8858850
calculate Weighted average SCF for maintenance cost (labour
economic cost) and material) 0.8456175
Source: LGED (internal communication with the project team)
4
LGED has established design and construction standards based on defined traffic volumes.
Page50
With the increase of traffic over the years, the savings in travel time would decrease as the average
travel speed would gradually fall. In less than 10 years the volume of traffic on many rural roads is
expected to double and may exceed the present capacity of roads. Consequently, beyond 10 years,
it is unlikely that there would be any significant savings in travel time. Therefore, for the purpose of
this study, after 10 years no additional benefit from savings in travel time has been considered.
The improvement of road roughness condition results in higher speed and thereby reduce travel
time. The benefits from reduction in travel time or in other words savings in travel time have been
estimated and used in the analysis. Standard values of VOCs by vehicle type, average speed of travel
under different road surface roughness condition, and value of time savings for both passengers and
vehicles have been considered from a study undertaken by GTZ for LGED (GTZ, 2009). It may be
mentioned here that in determining the average passenger time saving values the GTZ study
considered the split between work and non-work trips. The parameter values taken from the GTZ
study were converted into current prices by using the consumer price index or CPI value. The VOC
values can also be updated in the software when more recent values are available.
It is important to note here that all or some of these parameter values can be changed before
running the software, should there be necessity, for example, in another district where construction
and maintenance costs are different, or when the tax rate becomes different after few years.
In estimating benefits, benefits to generated traffic have been estimated at half the benefits to
current traffic.
For this study, benefits from any potential increase in agricultural production have not been
considered for two main reasons. First, road improvement is a necessary condition but not sufficient
to generate new development/production activities. Any potential increase in agricultural
production (or any other type of economic activity) cannot be solely attributed to improvement of
road. Other complementary interventions (such as investment for such increase in production,
development of transport services and logistics) may also be necessary. Second, the demand for
transport is primarily a derived demand, meaning that the benefits from improved transport are
passed through to prices for products and factors. Separate consideration of such benefits would
have led to double counting of benefits, at least to some extent. Since additional production/
economic activities produces “generated traffic”, benefits from which have already been considered
in the analysis.
Page51
Table 3.12: Vehicle operating cost and speed by road surface condition, passenger time cost, average occupancy and PCU factor by type of vehicle
VOC VOC
Good (IRI VOC Fair Poor (IRI VOC Bad
less than (IRI in in (IRI
Vehicle or equal between between greater Speed_ Speed_ Speed_ Speed_ Average PCU_
ID Vehicle Name to 6) 6-8) 8-10) than 10) Good Fair Poor Bad TTS Occupancy Factor
1 Motorcycle 3.65 3.69 3.71 4.14 79.00 67.00 62.00 44.00 22.90 2.00 0.30
2 Tempo 5.65 5.73 6.51 6.95 0.00 0.00 0.00 0.00 16.40 10.29 1.00
3 Car 19.49 19.66 21.14 26.20 81.00 77.00 66.00 44.00 30.90 3.52 1.00
4 Delivery 13.45 14.09 15.97 19.30 0.00 0.00 0.00 0.00 27.36 3.48 1.00
5 Minibus 34.12 34.39 37.10 48.26 69.00 68.00 62.00 43.00 17.60 35.38 3.00
6 Medium Truck 26.39 26.85 29.10 36.31 59.00 57.00 52.00 40.00 32.36 3.24 3.60
7 Bicycle 1.97 2.03 2.38 3.34 N/A N/A N/A N/A 12.28 1.22 0.30
8 Animal Cart 9.68 10.21 11.29 15.81 N/A N/A N/A N/A 0.00 0.00 4.00
9 Rickshaw 2.86 3.06 3.86 5.54 N/A N/A N/A N/A 12.28 2.32 1.00
10 Rickshaw Van 2.86 3.06 3.86 5.54 N/A N/A N/A N/A 12.28 6.99 1.00
11 Small Truck 13.45 14.09 15.97 19.30 67.00 65.00 59.00 43.00 27.36 3.48 3.60
12 Large Bus 81.63 81.48 85.78 93.48 73.00 70.00 59.00 39.00 17.60 46.59 3.00
13 Light Bus 18.27 18.35 19.68 25.14 69.00 68.00 62.00 44.00 17.60 16.18 3.00
14 Utility 19.32 19.43 22.85 30.46 82.00 77.00 65.00 43.00 30.90 6.97 1.00
15 Auto-rickshaw 5.65 5.73 6.51 6.95 48.00 47.00 44.00 36.00 16.40 4.00 1.00
Source: Extracted from GTZ (2009).
Page52
3.7.3. Limitations of the adopted CBA methodology
The CBA methodology developed for the purpose of this study is a simplified one. It does not
require collection of a vast amount of data for individual roads. The methodology is based on unit
cost/standard values, for example unit construction and maintenance costs and standard conversion
factors for shadow pricing to estimate economic costs. It is important to mention here that an
average upazila has some 400-600 rural roads. It is not practical to collect/estimate specific data for
such a large number of roads to undertake CBAs. The methodology, therefore, mainly relies on data
available from the road database, and LGED and other secondary sources. However, the adopted
methodology follows the standard procedures in estimating economic costs and benefits of road
improvement as far as practical.
3.8. Salient features of the methodology
There are several important features of the methodology. The main ones are briefly discussed here.
There is a significant research component in developing the methodology to establish many
parameter values that have used in both CBA and MCA, as well as road identification for their
inclusion in different types of road development. For example, the weights assigned to socio-
economic facilities were established by applying the Reed-Muench ‘Median Threshold Population’
method based on the data from a purposely designed field study.
The conventional approach to network planning focuses on traffic flow. The main purpose is to
minimise travel time and/ortravel distance or maximise traffic flow capacity. Methodologies such as
the Minimal Spanning Tree may also be considered to minimise network length. However, the
network planning approach adopted for the present study focuses on connectivity to ensure
accessibility, the primary objective of rural road development, at all levels – village, union and
upazila. The activity centres (upazila and union headquarters, and growth and market centres) form
the nodes, and upazila and union roads and some village roads connecting the nodes are the links of
a core road network. Villages are to be connected either directly to such centres or to a link of the
core network providing access to the facilities at those centres and/or along the links. The approach
also consider connectivity with the neighbouring upazilas as well as with the wider region. Thus, it
ensures access for the rural people to facilities both within and outside an upazila.
The methodology for the study combines both the top-down and bottom-up approaches to
planning. While the adopted methodology reliesmainly on a rational approach to planning and
prioritisation that considers an objective evaluation of planning factors, it also considers valuable
inputs from local workshops to improve the overall process and the quality of output. The local
workshops complement the technical process, and to some extent helps to overcome the limitations
of data available from the current road database. Such workshops also help to show transparency in
the planning and prioritisation process and ensure local ownership of the output.
LGED undertakes three types of rural road development activities: improvement, further
improvement and maintenance. All of these activities have been considered within a single planning
framework based on priority evaluation of roads following a consistent approach.
The methodology is centred on the idea of developing a software that is easy to use, makes best use
of the available data from the databases and can enhance decision making in LGED. It can also help
in project formulation and road master plan preparation.
Page53
3.9. Conclusion
Some of the major considerations in developing the methodology for this study were that it should
not require any vast amount of new data collection, practical and can be easily used for other
districts in Bangladesh, and meets the requirements of LGED. The project team developed the
methodology in direct collaboration with the concerned LGED officials to meet these requirements.
As described in the preceding sections, the developed methodology meets all these requirements.
So far the methodology is concerned, apart from periodic updating of data in the databases, only
with some parameter/unit values which will require updating time to time when such new values
become available. These values can be easily changed before running the software (RPPM) which
has been developed to implement the methodology.
The methodology also meets the strong demand of local stakeholders for their involvement in the
planning and prioritisation process. It allows their participation in the process through local
workshops. Some valuable inputs from them are collected to validate/finalise the core road
networks and prioritise road development.The local workshops complement the technical process
and also can ensuretransparency in the planning and prioritisation process of future road
development works by LGED.
Page54
4 Tangail district: The study area
4.1. An overview of the district
Tangail district was selected as the study area for piloting the methodology. It is 13th largest district
of Bangladesh by area and 5thlargest by population. It is the largest district of Dhaka division by area
and 2ndlargest by population following Dhaka district. It lies between 24° 01′ and 24° 47′ north
latitudes and between 89° 44′ and 90° 18′ east longitudes. Figure 4.1 shows the administrative units
along with several important features of the district.
Figure 4.1: Map of Tangail district
Source: LGED, 2016
Page55
Tangail district is surrounded by Jamalpur district on the north, Dhaka and Manikganj districts on the
south, Mymensingh and Gazipur districts on the east, and the Jamuna River on the west (Figure 4.1).
The city of Tangail is located around 86 km North-west of Dhaka. The district is mostly crisscrossed
by roads. National Highway 4 (NH 4), which passes through this district, connects Dhaka with the
North-western part of Bangladesh through the Jamuna Bridge.
Per BBS (2011), population of Tangail district was 3,605,083 of which around 85% people live in rural
area. It encompasses 12 Upazilas (sub-districts) and nine paurashavas (municipalities). Ghatail is the
largest upazila by area; however, Tangail sadar stands out to be the largest upazila by population.
Table 4.1 provides some of the notable geographic and demographic information of Tangail district.
Table 4.1: An overview of Tangail district

Literacy
Area Population (2011) Density
Upa- Pauras rate (%)
(sq Union Mouza* Village (per sq
zila havas
km) Male Female km) M* F*
3,605,083 46.8
3,413.
12 9 110 1,884 2,443 1,847,71 1,056
68 1,757,370 50 44
3
*Mouza: Smallest Geographic Unit for Revenue Collection
*M: Male, *F: Female
Source: BBS, 2015
Table 4.2: An overview across the upazilas of Tangail district

Density
Area (Sq Population Literacy
Upazila Paurashavas Ward Unions Mouza Village (per sq
km) (2011) Rate (%)
km)
Basail 157.17 0 0 6 72 107 159,870 1,017 50.4
Bhuapur 225 1 9 6 81 128 189,913 844 43.8
Delduar 184.54 1 9 8 123 162 207,278 1,123 51.6
Dhanbari 133.75 1 9 7 103 132 176,068 1,316 44
Ghatail 451.3 1 9 12 291 411 417,939 926 44
Gopalpur 193.37 1 9 7 111 158 252,331 1,305 45.7
Kalihati 295.6 1 9 14 242 298 410,293 1,388 42.4
Madhupur 366.92 1 9 6 111 180 296,729 809 41.2
Mirzapur 373.88 1 9 14 202 210 407,781 1,091 55.5
Nagarpur 262.7 1 9 12 213 245 288,092 1,097 42.7
Sakhipur 435.19 1 9 6 59 122 277,685 638 41.1
TangailSadar 334.26 1 18 12 247 276 521,104 1,559 53.1
Source: BBS, 2012; BBS, 2013
Tangail district consists of 12 Upazilas, namely –Basail, Bhuapur, Delduar, Dhanbari, Ghatail,
Gopalpur, Kalihati, Madhupur, Mirzapur, Nagarpur, Sakhipur, and Tangail Sadar (Figure 4.2). Table
Page56
4.3 offers a basic overview with some demographic statistics across all the upazilas. Also, the
distribution of growth centres, and markets across the Upazilas of the study area is shown in table
4.4.
Figure 4.2: Upazilas of Tangail district
Being the administrative hub of the district, Tangail Sadar Upazila accommodates the highest
population of the study area followed by Kalihati and Mirzapur Upazilas. The average literacy rate
stands out to be around 46% in the study area which is about 15% lower compared to the national
average. Being a rural area, economic establishments like growth centres and markets partly dictate
the mobility pattern of the people. Though Tangail Sadar appears to be the most densely populated
Upazila but most of the economic establishments are located in Kalihati Upazila followed by
Nagarpur and Mirzapur Upazilas.
4.2. Growth centres, educational and other facilities in Tangail district
There are several growth centres, markets (locally known as hat or bazaar) and other activity centres
in the study area. Table 4.2 shows the number of these market centres and other activities
locatedhere.
Page57
Table 4.3: Number of growth centres, educational and other facilities in Tangail district
Facilities Number Facilities Number
Growth centre 77 Homeopathy college 1
Market 457 Madrasas* 202
Government primary schools 941 Technical university 1
Non-government primary schools 395 Nursing institutes 2
Community primary schools 146 Polytechnic institute 1
NGO operated schools 1,304 Medical assistant training school 1
Junior schools 40 Police academy 1
Satellite schools 86 Teachers' training school 1
Government high schools 5 Mosque 5,763
Non-government high schools 341 Temple 759
Government colleges 5 Church 56
Non-government colleges 48 Fuel Filling station 44
University colleges 3 Fire station 2
Textile engineering colleges 2 Police station 23
Medical college 1 Cyber cafe 9
Law college 1 Flood and Cyclone Shelter 18
*Madrasa: An educational institution for the study of Islamic Theology and Religious Law
Source: BBS, 2012; BBS 2013
Table 4 .4: Upazilawise distribution of growth centres and markets

Upazila Growth Centre Market Total Economic Establishment
Basail 4 36 40
Bhuapur 2 41 43
Delduar 5 16 21
Dhanbari 3 15 18
Ghatail 8 31 39
Gopalpur 4 33 37
Kalihati 6 72 78
Madhupur 12 21 33
Mirzapur 4 51 55
Nagarpur 3 74 77
Sakhipur 5 44 49
Tangail Sadar 21 23 44
Tangail District 77 457 534
Source: BBS, 2012; BBS 2013
Page58
4.3. Rural roads in Tangail
The length of total road network in Tangail district is about 8,250 km. The distribution of roads by
upazila is shown in Table 4.5. Road density (road length/s. km) in Tangail (2.42 km/sq km) is higher
than the national average (2.18 km/sq km). However, there is wide variation of road density by
upazila.The highest road density is in Nagarpur (3.17 km/sq km) and the lowest is in Tangail Sadar
(1.55 km/sq km).
Fig 4.3 (a) depicts that village roads (VR), with a split of 41% and 33% between VR Type A and VR
Type B, constitute the highest share of roads in Tangail district followed by Union road (about 14%)
and Upazila road (about 11%). This is essentially complemented by figure 4.2 (b) which portrays the
fact that most of the roads of Tangail district, about 78% of the entire network, are earthen. Of the
total unpaved or earthen roads, the shares of Upazila and Union roads are 3.49% and 10.2%,
respectively.
Figure 4.3 (a): Share of road length (km) by road type Figure 4.3 (b): Share of unpaved (earthen) road
length (km) by road type
Source for figures 4.3 (a) and (b): Data presented in Table 4.5
Figure 4.4 (a) – (d) offer a closer look into the share of road surface type by Upazila roads, Union
roads, VR Type A, and VR Type B. Based on the share of earthen roads across various administrative
levels, it is evident from the figures that surface type keeps deteriorating as the administrative
hierarchy moves down from Upazila to Village level. Also, share of paved roads dominates only for
Upazila roads (about 74%) followed by a notable decrease of Union roads (about 41%), and a
notable decrease for Village roads - about 12% and 4% for VR Type A and VR Type B respectively.
Page59
Figure 4.4(a): Share of surface type of Figure 4.4(b): Share of Surface type of Union
Upazilaroads roads
Figure 4.4(c): Share of surface type of Village Figure 4.4(d): Share of surface type of Village
road type A roads road type B roads
Source for figures 4.4 (a), (b), (c) and (d): Data presented in Table 4.5
Page60
Table 4.5: Road length and density in Tangail by upazila
Upazila Road length in km by type of road Area Length of

Name (length of paved road) (square road per
Upazila Union VR Type A VR Type B Total km) square
Road Road km
Basail 60.99 95.9 133.38 199.79 490.06 157.78 3.11
(31.30) (36.57) (16.45) (6.41) (90.73)
Bhuapur 25.8 49.64 196.88 160.53 432.85 225.02 1.92
(22.31) (30.19) (40.47) (1.30) (94.27)
Delduar 64.18 77.33 217.35 167.49 526.34 184.54 2.85
(43.88) (19.99) (12.91) (5.78) (82.56)
Dhanbari 41.99 79.56 201.38 41.28 364.21 130.5 2.79
(36.19) (45.31) (16.46) (7.40) (105.36)
Ghatail 102.48 117.67 549.37 338.25 1107.76 451.3 2.45
(76.39) (50.13) (77.61) (10.11) (214.24)
Gopalpur 73.6 49.56 179.79 263.91 566.85 193.37 2.93
(67.90) (42.43) (55.18) (22.97) (188.48)
Kalihati 126.52 124.84 349.01 210.73 811.09 301.22 2.69
(100.15) (50.16) (36.72) (11.97) (199.00)
Madhupur 52.79 99.94 331.09 278.95 762.77 366.92 2.08
(52.79) (53.08) (60.33) (19.79) (185.99)
Mirzapur 101.59 120.55 337.71 495.19 1055.04 373.89 2.82
(68.50) (45.16) (53.67) (30.42) (197.75)
Nagarpur 100.36 95.65 365.7 270.3 832.01 262.7 3.17
(78.58) (26.16) (18.83) (3.47) (127.04)
Shakhipur 108.79 140.56 307.89 225.92 783.24 429.63 1.82
(77.19) (38.18) (48.80) (21.29) (185.47)
Tangail 81.3 116.34 222.14 97.27 517.05 334.26 1.55
Sadar (61.75) (73.99) (37.49) (0.97) (174.19)
Total for 940.39 1167.54 3391.69 2749.61 8249.27 3411.13 2.42
Tangail (693.83) (478.63) (416.86) (105.29) (1694.61)
Total for 37399 41621 120899 121543 321462 147570 2.18
Bangladesh (28363) (19111) (19455) (7166) (74095)
Source: LGED road database and BBS, 2011
Page61
4.4. Traffic volume
Figure 4.5 shows the share of traffic volume, with a split between motorised and non-motorised traffic,
across different types of roads in Tangail district. It is evident from the figure that traffic composition is
very similar for all categories of roads with the predominance of motorised traffic on the Upazila and
Union roads (more than 50% of the traffic volume). However, non-motorised traffic takes a slight lead of
the traffic share with approximately 60% contribution to the total traffic volume on village roads.
Figure 4.5: Share of Traffic Volume across Different Types of Roads of Tangail
Source: Based on AADT data in the road database of LGED
Figures 4.6 and 4.7are constructed through generating a motorised and non-motorised traffic volume
analysis across different types of roads prevalent in the study area. It appears from figure 4.6 that
motorised traffic volume is lower in Village roads as opposed to Union roads and Upazila roads. As the
traffic volume exceeds 500 (AADT), distribution of motorised traffic volume leans more towards Union
and Upazila roads.Upazilaroads, as opposed to Union roads and Village roads, are the ones most
congested by motorised traffic. The higher the volume of motorised traffic, the higher the usage of
Upazila road. In contrast, village roads for most part hold up for low volume motorised traffic. Table 4.5
offers a detailed distribution of motorised traffic volume.
Page62
Figure 4.6: Distribution of motorised traffic volume across different roads of Tangail
Source: Based on AADT data in the road database of LGED.
Like table 4.5, it can be seen from table 4.6 that less than fifty roads in Tangail are experiencing
relatively high-volume of traffic (AADT > 600). It appears from figure 4.7 that Upazila roads are still at
their peak when it comes to percentage of roads being observed with relatively higher volume of non-
motorised traffic leaving both the village roads and union roads behind. In contrast, village roads,
followed by Union roads and Upazila roads, dictate majority of the roads categorised under
comparatively lower volume of non-motorised traffic.
Tangail being a regionally important district that bridges between the capital with the northern part of
Bangladesh is mostly a rural area. Over time, the spatial distribution of growth centres and markets have
paved the way for developing an organic, criss-crossed network of road infrastructure that allows both
motorised and non-motorised traffic to move around places. However, only a small share of roads is
experiencing higher volume of motorised and non-motorised traffic. Though the majority of roads
mostly carry a large share of low volume traffic but, especially, Village roads outweigh the other two
types of roads by a huge margin. Most of theUpazila roads are, in most cases, have higher volume of
motorised and non-motorised traffic.
Page63
Table 4.6: Distribution of motorised traffic volume across different roads of Tangail
Motorised Traffic Volume (AADT)
Frequency
Village Road Union Road Upazila Road
1 to 50 555 7 0
51 to 100 332 10 2
101 to 150 205 12 0
151 to 200 109 13 4
201 to 250 57 9 4
251 to 300 50 9 5
301 to 350 49 10 1
351 to 400 26 11 4
401 to 450 13 4 4
451 to 500 19 3 7
501 to 550 8 5 4
551 to 600 8 8 9
601 to 650 9 7 3
651 to 700 7 9 4
701 to 750 4 9 3
751 to 800 3 8 4
801 to 850 0 6 2
851 to 900 3 8 3
901 to 950 0 6 5
951 to 1000 0 8 2
1001 to 1050 0 0 5
1051 to 1100 1 0 0
1101 to 1150 0 5 2
1151 to 1200 1 2 1
>1200 1 3 26
Source: Data from the road database of LGED.
Figure 4.7: Distribution of non-motorised traffic volume across different roads of Tangail
Page64
Table 4.7: Distribution of non-motorised traffic volume across different roads of Tangail
Frequency
Non-motorised Traffic Volume (AADT) Village Road Union Road Upazila Road
1 to 50 326 1 1
51 to 100 311 3 1
101 to 150 299 8 2
151 to 200 195 13 0
201 to 250 141 14 1
251 to 300 94 15 7
301 to 350 58 17 7
351 to 400 37 17 13
401 to 450 32 18 10
451 to 500 28 9 8
501 to 550 20 8 4
551 to 600 13 8 5
601 to 650 3 5 6
651 to 700 2 7 4
701 to 750 3 3 5
751 to 800 1 4 3
801 to 850 1 0 2
851 to 900 1 1 2
901 to 950 1 2 0
951 to 1000 3 1 2
1001 to 1050 0 1 4
1051 to 1100 1 0 2
1101 to 1150 0 3 3
1151 to 1200 2 3 0
>1200 0 4 12
Source: Data from the road database of LGED.
Page65
5 Local level workshops
5.1. Introduction
During the First Stakeholders Workshop, held at the LGED Head Quarter on the 9 August, 2016, local
representatives including a local Member of Parliament (MP) and Upazila Chairmen raised the issue of
participation of the local people in rural road development process. In response to their demand Local
Workshops were designed to capture the views of the local stakeholders related to planning and
prioritisation of rural roads in their locality.
Local Level Workshops have an important role in the implementation of the methodology developed for
planning and prioritisation of rural roads. The feedbacks received from the twelve workshops done at
twelve upazilas of Tangail districts were crucial to validate core network and collect data for the
prioritisation methodologies. The prioritisation software developed for this project used data collected
from these local level workshops. Most of the data required to implement the methodology and
consequently the software are available in LGED’s Road Data Base Management System (RDBMS).
However, some data that were not available and hence had to be collected from the field through these
workshops.
5.2. Objectives of the local level workshops
Local level workshops were organised with specific objectives. The objectives were as follows:
- To facilitate participation of local leaders and other stakeholders in rural road planning and
prioritisation process.
- To verify the Preliminary Core Network, generated by using information available in the RDBMS,
by the local stakeholders and identify village roads, if any, have the merits to be part of the core
network.
- To identify the local priority of all the rural roads.
- To collect information on the type of further development required for paved roads and road
safety condition of these paved roads.
The local level workshops complement the technical process of planning and prioritisation of rural
roads. These workshops also helped to show transparency in the decision making process, ensure local
ownership of the output and overcome the limitations of data available in LGED’s RDBMS and GIS
platform.
5.3. Organisation of the local level workshops
After finalising the methodology, the BUET team generated the preliminary core network for each
Upazila using the GIS data and RDBMS of LGED. The Executive Engineer of Tangail district was
communicated through our counterpart at LGED Headquarter at least two weeks before a proposed
date for the workshop at each upazila. The Executive Engineer consulted the respective Upazila Engineer
- who subsequently contacted Upazila Chairman and Upazila Nirbahi Officer (UNO) to check the
availability of local stakeholders on the proposed date. Otherwise they proposed an alternative date.
Page66
When the date was fixed, Upazila Engineer was informed about the logistic support required. Upazila
Engineer, in consultation with Upazila Chairman, invited the participants for the local level workshop.
Upazila Engineer and Sub-Assistant Engineers at each upazila were appraised in detail the purpose of the
Local Level Workshop and tasks to be done during the workshop. In some cases they provided context
specific suggestions for the successful implementation of the workshop at Upazila levels.
During the Workshop day, the BUET team used to reach the venue at least one hour before the starting
of the workshop. Members of the BUET team made courtesy calls with the Upazila Chairman and UNO.
Meanwhile, the venue was prepared and the programme started after the arrival of invited participants.
It took about three hours to complete each of the Local Level Workshops.
In order to facilitate discussion in the two technical sessions with local stakeholders some logistics
support was required. So, the BUET team arranged all the required documents in advance and put those
in front of the concerned persons. The language of the workshops was Bengali. It was helpful for the
participants to follow the visual presentations by the BUET team and following discussion. Thus care and
considerations were made so that stakeholders at local level could discuss and converge to some
decisions regarding rural road planning.
5.4. Participants of the local level workshops
Participants of the local level workshops were local people who had interest in rural road development
as decision makers or users. Follows groups of participants were invited to attend and participate at the
workshop: local MP, Upazila Parishad Chairmen, Vice Chairmen (particularly Women Vice Chairmen),
Union Parishad Chairmen and members (particularly women members), Upazila Nirbahi Officers (UNO)-
Chief Executive Officers at Upazila level, Upazila Engineers and Sub-assistant Engineers of LGED,
Executive Engineer of LGED at Tangail District, Officer-in-Charge of the local Police Station, officials of
the upazila administration, Members of Nagorik Committee (local chapter of a national citizens right
organisation) and others. On an average, twenty to twenty five people were present in the workshops.
5.5. Activities undertaken at local level workshops
Four types of activities held in two sessions. They were as follows: inauguration of the workshop in the
first session; core network map verification, local priority ranking of rural roads and identification paved
roads for further improvement in the next session. The first session of the workshop was moderated by
the Upazila Engineer of LGED and the other tasks in the next session were conducted by the BUET team.
In the first session Upazila Engineer welcomed the participants in the workshop. The special guests –
senior most representative of the BUET team, Upazila chairman, Member or the Parliament and Upazila
Chairman addressed the participants subsequently. They underscored the importance of the workshop
in their speech. The BUET representative explained the projects in brief and the objectives of the project
and workshop. Finally Executive Engineer, Chair of the inaugural session, requested all the participants
to take part in the workshop actively and help LGED in undertaking priority wise road development
schemes based on merits. It took about half an hour to complete the session.
Page67
In the next sessions, the participants were requested to complete three main tasks. They were as
follows:
- Validation of the Preliminary Core Network
- Consideration of local priority for all the roads in the Upazila
- Identification of potential roads for further improvement
At first a BUET team member explained the concept of Core Network and showed the participants the
preliminary core network for respective Upazila in an A0 (33.1X46.8 square inch or 841X1189 sq.mm)
size printed map. All the participants were given A4 size printed copies of the core network maps. Core
network maps for the Upazila and for the respective unions were printed separately and were
distributed to the concerned participants. At some Workshops the preliminary core networks drawn by
the BUET team were verified by the local stakeholders to be complete and they added no new village
road into the network.
However, in some upazilas, local stakeholders, particularly Union Parishad members, proposed some
potential village roads to be included. The other stakeholders, including local Engineers from LGED,
further examined the proposal. Finally, some of the proposed village roads were included and rest were
not. As the local people know the best which village road had important new establishments
(considered in the definition of core network) on both sides and hence had the merit of being included
into the network, without consulting them the Preliminary Core Network could not be finalised. Thus
after the local level workshops and inclusion of new village roads, if any, validation of preliminary
network was completed.
After validation of core networks, feedback on local priority for each of the roads in the Upazila was
collected. Upazila Chairman and Vice Chairmen were given the list of all Upazila roads and union roads
in the Upazila; the Union Parishad members were given the list of village roads in their respective
unions. They were requested to put priority values (3 for highest, 2 for Medium and 1 for lowest
priority) for each of the roads in the list based on their perception and local knowledge and experience –
no criterion or condition was set
Finally, to accomplish the third task, BUET team requested all the local stakeholders to select paved
roads in the upazila requiring further improvement. The Upazila Engineer and Sub Assistant Engineers of
LGED gave feedback on types of further improvement (widening, raising elevation, alignment
straitening, pavement strengthening, etc.) required for the selected paved roads. They, Upazila and Sub-
Assistant Engineers, also commented if any of those selected paved roads had regular record of
accidents or not. This information is included in the database used by the BUET team. It is expected that
in future this information will help the decision makers to take more informed decisions and to compare
their decision with local engineers’ requirements.
5.6. Workshop materials distributed
While conducting the workshops potentially required materials were distributed among the
participants. It should be mentioned that all stakeholders were not given same materials as all of them
were not required to provide the same feedback. For example, representatives of Union Parishads were
concerned with union-wise core network maps and priority ranking of village roads in their respective
Page68
unions, while the representatives from the Upazila Parishads were concerned with the core network
map at Upazila level and priority ranking of Upazila and Union roads.
The following materials were used to conduct the local level workshops:
- Printed copies of updated list of union wise list of rural roads. At least two copies for each union
were required.
- Printed copies of list of upazila and union roads. At least three copies were required
- Printed copies of Preliminary Core Network map for the entire Upazila. At least one A0 size map
and 20 A4 size printed maps were required.
- Printed copies of Preliminary Core Network maps for each union. At least two union-wise
Preliminary Network maps for each union were required.
Upazila and Union level participants were requested in advance to bring their official stamp. They signed
and stamped the prioritised list of roads and map of the validated core network.
5.7. Logistics needed
To conduct a workshop at Upazila level a venue with modest sitting (and writing) arrangements for at
least thirty persons were required. Upazila Parishad Meeting Room or Muktijodhdha Auditorium in the
Upazila premise was found suitable to serve the need of the workshop. In upazilas where computer and
projector could be used, the deliberations by the BUET team were easier for the participants to
understand and follow. Prior arrangements were also made to hang the large size maps of preliminary
core network and banner for the program. In some cases alternative power supply arrangements and
availability of printing and photocopying facilities were useful.
5.8. Outcome of the local level workshop
The following were the main outcome of the workshop:

 Verified upazila and union core road networks
 Identified upazila, union and village roads that have local priorities for development
 Identified roads that may need further development (road widening, improvement of road
geometry and road safety features etc.) and
 Collected local inputs and data necessary for CBA and MCA analyses
The outcomes were used to update the RDBMS. New fields were created in the database to incorporate
the data. The planning and prioritisation methodology could be executed with all the relevant inputs.
5.9. Challenges faced
In most cases the workshops were held smoothly. However, problems arose when some stakeholders
had difficulties in locating roads of their interest in the list provided. It happened either due to their
difficulties in reading maps or absence of that road in the LGED GIS map or database. The BUET team
and LGED Upazila staff helped them to find the roads. They were advised to inform the Upazila Engineer
when road database would be updated next time. LGED Upazila staffs also took note of the omission so
that during the next database updating the road could be included.
Page69
5.10. Post workshop activities
After the workshop, the BUET team visited some local roads and observed their conditions. The roads
included upazila roads, union roads and suggested village roads, by the local stakeholder, in the core
network. The team tried to understand the actual condition of the roads and variation in their priorities
provided by the local stakeholders.
5.11. Summary of local level workshops held in the Tangail district
A total of twelve local level workshops were held in the district between the month of November, 2016
and February, 2017. The table below provides a summary of local level workshops held Tangail.
Table 5.1: Summary of local level workshops at Tangail district

Workshop Upazila Date Number of Remarks/Participants
No. Participants including
1 Tangail Sadar November 24, 2016 26 Local Member of
Parliament and
Executive Engineer ,
Tangail
2 Delduar December 27, 2016 24 -
3 Dhanbari December 28, 2016 20 -
4 Basail January 12, 2017 21 Representative of MP
and President, Nagorik
Committee
5 Shakhipur January 12, 2017 19 -
6 Mirzapur January 17, 2017 30 Executive Engineer,
Tangail
7 Nagarpur February 1, 2017 28 Local Member of
Parliament
8 Bhuapur February 1, 2017 18 -
9 Kalihati February 1, 2017 27 -
10 Ghatail February 2, 2017 23 -
11 Gopalpur February 2, 2017 21 -
12 Modhupur February 2, 2017 14 -
Page70
6 Rural road planning and prioritisation software
6.1. Introduction
One of the objectives of the project is to develop a computer based tool on prioritisation of
development and maintenance of rural roads in Bangladesh. Accordingly, the project has developed the
Rural Road Planning and Prioritisation Model (RPPM), a web-based software to implement the planning
and prioritisation methodology discussed in Chapter 3. RPPM has been developed to meet the
requirements of LGED for prioritisation of rural roads for “improvement”, “further improvement” or
“maintenance” as may be required. The definitions of these terms are provided in section 1.3.
The software was developed using C# (C sharp) program. RPPM works in the web platform and can
return output data in both tabular (report) and graphical (map) form. At present RPPM can produce
outputs for 12 upazilas of the pilot district Tangail but can be applied to generate similar outputs for
other districts of Bangladesh.
It is expected that in the future RPPM would be used by the Planning, Design and Procurement, and
Maintenance, Asset Management and Road Safety departments of LGED to produce prioritised lists for
rural road development for other districts. A user with access to the Internet can use the software by
logging into the GIS portal of LGED. RPPM performs the following tasks:
 Prepare prioritised lists of different types roads for improvement both in tabular and
graphical forms
 Prepare prioritised lists of different types of roads for further improvement both in
tabular and graphical forms
 Prepare prioritised lists of different types of roads for maintenance both in tabular and
graphical forms
 Generate a score table with basic information, details of priority scores, and CBA and
MCA results of a road, and generate the Core Road Network of an upazila in graphical
form.
6.2. Structure of RPPM
RPPM consists of three different subsystems. Figure 6.1 shows the schematic structure of RPPM
showing the functional relationship of the three subsystems. However, it does not show the actual flow
of data which can be found in Figure 3.3. One of the three sub-systems generates the core road
network. Three interconnected modules form the second sub-system which uses the evaluation models
for this study (MCA and CBA) to produce the prioritisation lists of roads for improvement, further
improvement and maintenance. The third sub-system is the output subsystem which shows the results
in the required format. The first two sub-systems work in the back end of the software and is not visible
to the user.
Page71
Figure 6.1: Schematic structure of the software
LGED DATA Base
Module for Traffic

Projection Module for Road Selection
LGED Core Road
Network
Module for VOC
Calculation
Module for CBA Module for MCA
Calculation
Module for TTS
Calculation
Module for EIRR Prioritized list of Roads for

Calculation Prioritized list of Roads
Improvement and Further
for Maintenance
Improvement
Map of Prioritized list of Map showing Prioritized list of

Roads for Improvement Road for Maintenance
and Further Improvement
Map of LGED Core

Road Network LGED GIS
Page72
The sub-systems (other than sub-systems for output) use data from different tables of the road
database. Using the data from the database, it determines the eligibility of roads in the database for
different types of development (improvement, further improvement and maintenance), as well as if a
particular road can be considered to be a part of the core network. Later, RPPM determines the priority
ranking of each road by considering the results of CBA and MCA analyses. The output module integrates
the GIS database with the results generated by the subsystems.
6.2.1. Subsystem for generating core network

This subsystem generates core road network for the selected upazila. Using the following criteria it
generates core network.
 All Upazila roads would be part of core network
 All Union Roads would be part of core network
 Village roads having the facility score above 50 in a scale of 100 would be part of core
network
It may be mentioned here that in an upazila if there is any national and regional highways or zila roads,
which are managed by RHD, by definition they will also be part of the upazila core network as given
roads.
The LGED road database do not have data on any roads of RHD. However, the GIS database of LGED
maintains the RHD network on a separate layer. As such, the RHD roads cannot be shown in the
generated core road network map (Fig 6.2).The complete core network of an upazila with RHD roads, if
any, can be seen when the RHD road layer is activated with the core network output generated by this
subsystem. The output of this subsystem is available only in graphical form.
Figure 6.2: Output of core road network sub-system
Page73
6.2.2. Subsystem for prioritisation
This sub-system generates the lists of prioritised roads for different kinds of development. It consists of
three independent but interconnected modules. The modules are:
 Road Selection Module for particular type of Development
 Module for Multi-criteria Analysis (MCA)
 Module for Cost-Benefit Analysis (CBA)
The road selection module is the module that selects roads for improvement, further improvement and
maintenance. This module determines the type of development a particular type of road may be
suitable for. Earthen, partially paved and HBB roads are selected for improvement; full BC/RCC and
HB+BC/RCC roads are selected for further improvement; and fully paved, partially paved and fully HBB
are selected for maintenance.
The MCA (Multi-criteria Analysis) module calculates MCA score for all roads. The module uses data on
volume of traffic, number and type of socio-economic facilities including growth centre and rural
markets, surface type and condition in different sections of a road, whether the road is a bus route,
connectivity and local priority to calculate MCA score. With the exception of connectivity and local
priority all the data are available in different tables of the existing road database. For connectivity and
local priority a new table fielddata is added in the existing database. The module extracts the required
data from different tables of the road database and then calculates the MCA score using the
methodology described in section 3.6 of this report.
The Cost Benefit Analysis or CBA module undertakes the economic evaluation of investment on road
development. The module calculates the Economic Internal Rate of Return of investment for
improvement or further improvement of roads. The CBA module of RPPM has four sub modules:
 Sub-Module for traffic projection
 Sub-module for calculating Vehicle Operating Cost (VOC)
 Sub-module for calculating benefits from Travel Time Saving (TTS),and
 Sub-module for calculating EIRR (Economic Internal Rate of Return).
A new table (Table 3.12) VehicleAttribute was added to the LGED road database to meet the data
requirements of the sub-modules (i.e. Sub-module for VOC, Sub-module for TTC and sub-module for
EIRR) of the module. The VehicleAttribute table also contains data on vehicle occupancy and PCU values
for different types of vehicles. These values are required for CBA.
Sub-Module for projecting traffic volume forecasts the future traffic generated due to improvement and
further improvement of the road. It projects both motorised and non-motorised traffic. In projecting the
future traffic 12% (6% normal traffic growth + 6% generated traffic growth) per annum growth rate is
considered for the first 10 years. For the next 10 years only normal traffic growth is considered (refer to
section 3.7 for details). The sub-module also uses data from other tables of the road database.
The Sub-module for VOC calculates the benefits derived from reducing vehicle operating cost due to
improvement of roads. Considering the roughness condition of a road, it calculates such benefits for 15
Page74
different types of vehicles which usually operate on rural roads. Similarly, the sub-module for TTS
calculates the time saving benefits to passengers for road improvement. It estimates travel time savings
to passengers for each type of vehicle and calculates the monetary value of such time savings
considering the value of passengers’ time for each category of vehicle.
The sub module for EIRR calculates the Economic Internal Rate of Return of road investment using the
internal output generated by the other three sub-modules. In order to calculate EIRR values, this sub-
module first converts the financial costs into economic costs by applying standard shadow price factors.
RPPM has the option to change the unit cost values by a user to reflect cost variation in different parts
of the country as well as cost in future years. The module returns the EIRR values considering 20 years of
economic life of a road. It should be mentioned here that at the development stage of the software,
results generated by all the modules and sub-modules of these subsystems were validated against
results produced by a separate spreadsheet model for CBA.
6.2.3. Subsystem for generating Output

The subsystem generates the final output of the exercise. Two types of outputs are generated – tabular
and graphical. Separate reports are generated in a tabular format with priority order of roads selected
for improvement, further improvement or maintenance. Eight different reports (Table 6.1) are
generated.
Table 6.1: List of reports generated by RPPM
Development Type Surface Type
Improvement Low Traffic Volume Earthen Roads
High Traffic Volume Earthen Roads
Partially Paved
Full HBB
Maintenance Partially Paved
Fully HBB
Fully Paved
Further Improvement Fully paved (Full BC/RCC, HBB + BC/RCC)
In the graphical mode, the selected roads for different types of development may be viewed on a map.
RPPM can show the roads for development for a single upazila or multiple upazilas. It is also possible to
look at the details of output (details of MCA and CBA results and other road attributes) of a selected
road on the map. The output module also shows the list of roads which constitute the core network of
an upazila in graphical form. As mentioned earlier, the LGED road database does not have information
on RHD road. So RHD roads are absent in the core network. However, by activating the RHD road layer,
which is available in GIS, the complete core road network for an upazila or selected multiple upazilas can
be viewed.
6.3. Data requirements
One of the conditions adopted during the development of the methodology was to keep the
requirement of additional data as little as possible. Most of the data required for the software are
collected through regular data collection and updating process of LGED. In addition to the data regularly
Page75
collected and available in the databases, RPPM needs some additional data. These include data on
connectivity, road safety and local priority. For Tangail, the project team collected these data from local
workshops. While the field offices of LGED can provide data on connectivity and road safety, the data on
local priority are to come from local workshops (see Chapter 5 for details). In addition, data on standard
construction and maintenance costs are required. The local/district offices of LGED can provide such
data. RPPM also needs data on vehicle operating costs under different road surface conditions, value of
passenger time, vehicle occupancy and PCU values. RPPM uses these data from an additional table
appended to the road database. The values in this table are extracted from a GTZ study for LGED.
6.4. Description of the software
RPPM is a web-based application. It has been integrated as a tool for analysis in the GIS web portal of
LGED. To use the application one must access the GIS web portal of LGED (http://www.gis.lged.gov.bd/).
At the time of writing the report, it is not yet generally accessible, as the portal itself is going through an
upgrading process. Figure 6.3 shows the home page of the portal. There are three options of which two
(Maps and Road Priority) are connected with RPPM.
Figure 6.3: Opening window of RPPM
As web base portal, the program must have the facility to control access to its database. It is understood
that for the time being RPPM would be run by the officials at the headquarters and if some parameter
values are needed to be changed they would do so. Some options though would be visible to a common
user, they would not be able to use it.
6.5. Programme initialisation
Once the data is ready for a district or an upazila (see section 6.7 for details on how to prepare and
update data), a responsible officer at the LGED headquarters can initiate the program. He can do so by
getting into the road priority mode of the GIS portal. Once inside this mode, a district and an upazila has
Page76
to be selected for running the process by clicking the start process (Figure 6.4a) and the program starts
running. One can run the process either for one upazila (Figure 6.4b) or for a whole district (Figure
6.4c).If there is a need to reconfigure the default parameter values, it can be done so by clicking
configure (Figure 6.5a). A pop-up window (Figure 6.5b) shows the current parameter values that can be
changed.
Figure 6.4a: Initialisation Windows of RPPM
Figure 6.4b: RPPM processing for an upazila
Figure 6.4c: RPPM processing for a district
Page77
Figure 6.5a: Configuring values for running RPPM
Button for
Configuring
Page78
Figure 6.5b: Pop-up window for changing values for EIRR
6.6. User interface

The output module generates results in two different modes. These are:
i. Reporting Mode
ii. Mapping Mode
Both of these modes are accessible to ordinary users. However LGED may also impose control on
accessing database or modification of values.
6.6.1. Reporting mode

In this mode a user views the priority list of roads for different types of development of an upazila in
tabular form. Once a user chooses this mode a page is displayed to give option to the user for selecting
upazila and types of development (Figure 6.6). For the time being, a user can only choose among the 12
upazilas of Tangail district. A user can specify a single upazila and then generate priority score tables for
that upazila.
Page79
Figure 6.6: Windows for choosing options for user
There are three options in the Reporting Mode:

i. Unit for generating list of roads for improvement
ii. Unit for generating list of roads for maintenance
iii. Unit for generating list of roads for further improvement
Page80
The unit for generating list of roads for improvement consists of four sub units. Initially the unit asks
(there is a drop down menu in the application) the user to identify the types of road for improvement.
Once the road type is selected (i.e. a sub unit selected) it generates a list containing the priority ranking
of the roads of that type (Figure 6.7a to 6.7e). The table shows the name and road code along with road
length, traffic volume data (both total and commercial) and MCA scores and EIRR values (where
applicable). For the earthen roads, with low volume of traffic, and roads for maintenance, ranking is
based on MCA scores. For the other categories of roads, ranking is done on the basis of EIRR values. If
there is no road for a particular type of development it let the user know that there is no road for that
kind of development (Figure 6.7e).
Figure 6.7a: A sample of output showing priority list for improvement for low volume earthen road
Page81
Figure 6.7b: A sample of output showing priority list for improvement for high volume earthen road
Figure 6.7c: A sample of output showing priority list for improvement for partially paved road
Page82
Figure 6.7d: A sample of output showing priority list for improvement for HBB road
Figure 6.7e: A Sample of output with no road to choose
Page83
Fully paved, partially paved and fully HBB roads are considered for maintenance. For this type of
development CBA is not undertaken. As such, ranking for prioritisation of road maintenance is based on
MCA scores. The priority list can be viewed (Figure 6.8a and 6.8b) by looking at different options under
the select option button. The generated lists provide road id, name of the road, total length, traffic
volume data (AADT, CVD), MCA score and ranking. Similar to improvement if no road is selected for
maintenance of a particular type of road, it would show there is no road for this category.
Figure 6.8a: A sample of output showing priority list for maintenance for partially paved road
Figure 6.8b: A sample of output showing priority List for maintenance for HBB road
Page84
Only one single list is generated for further improvement. Ranking is done on the basis of EIRR value.
The format of the table is the same as the format of the table that is generated for prioritisation of HBB/
partially paved road for improvement.
6.6.2. Saving the results
The lists generated in tabular mode can be saved (Figure 6.10) as a data sheet in spreadsheet (.xlsx), as a
report in portable document format (.pdf) or as a table in Microsoft Word document (.docx) format.
Using the home button (Marked in Figure 6.9) one can always return to the opening window.
Figure 6.9: Saving the Generated List
Home
Button
Saving the
Result
6.6.3. Graphical mode
A user can view the identified roads for a particular type development on a map. Unlike tabular mode, in
graphical mode the user can see all the identified roads of an Upazila (Figures 6.15 to Figure 6.20) or a
number of upazilas (Figure 6.12). It should be noted here that the tabular mode can only show one
particular development for one upazila. In addition, the core road network of an Upazila (Figure 6.3) or a
number of upazilas (Figure 6.11) can be viewed in the graphical mode. In this mode of RPPM, there are
many standard features like layers of map to select, combo box for administrative unit to view, zooming
options etc. (Figure 6.10). The RPPM tool is in the lower middle of the window (marked in Figure 6.10).
Page85
Figure 6.10: Opening screen of graphical mode
Combo Boxes for Selection
Zooming Tool
GIS Feature Layers
RPPM Tool
Figure 6.11: Core road network of a number of Upazilas (three)
Page86
In the graphical mode, a user can visualise the results of RPPM for a single or multiple upazila. However
for visual clarity, the examples shown in this report is for a single upazila.
Once the user activates the RPPM tool (Figure 6.12) from the graphical mode, it shows ten options for
road maps – one each for 8 types of road development, one for core network and none as the default
option. The core network of the upazila(s) may be viewed by selecting core road network (figure 6.13).
Once the National Road Network layer is activated in the left panel (marked in Figure 6.13) of the
window, the core road network of the upazila(s) including the given roads of RHD, if any, is shown
(Figure 6.14).
Figure 6.12: Activation of RPPM tool
Page87
Figure 6.13: Core road network of an upazila maintained by
LGED
RHD Road Layer
Figure 6.14: Core road network of an upazila including the RHD roads
Page88
Similar to the core network, a user can see the prioritised fully HBB road for improvement (Figure 6.15),
partially paved road for improvement (Figure 6.16), earthen road with low volume of traffic (Figure
6.17), earthen road with high volume of traffic (Figure 6.18), or further improvement of paved road. It is
also possible to view on a map the prioritised roads for other types (Figure 6.19 and Figure 6.20) of
developments.
Figure 6.15: Identified HBB road(s) of an upazila for improvement
Figure 6.16: Identified partially paved road(s) of an upazila for improvement
Page89
Figure 6.17: Identified earthen road(s) with low volume of traffic for improvement
Figure 6.18: Identified earthen road(s) with high volume of traffic for Improvement
Page90
Figure 6.19: Identified partially paved road(s) of an upazila for maintenance
Figure 6.20: Identified HBB road(s) of an upazila for maintenance
Page91
Instead of priority ranking, in the graphical mode the priority is shown by priority class. Depending upon
the number of candidate roads the number of class varies. However by double clicking the result layer
(shown in Figure 6.21a), a user can change the different parameters for display (Figure 6.21b)
Figure 6.21a: Identification of a layer for changing display
Resultant Layer
Figure 6.21b: Changing the display settings
In addition, a user can look at individual scores and ranking of any road selected for any type of
development. If a user selects a particular road, in a pop-up window it would provide a complete
Page92
description of priority score and ranking (Figure 6.22). These results can be saved in spreadsheet format
(*.xlsx), portable document format (*.pdf) or Word format (*.docx). It should be mentioned here that
the generated graphical outputs can be saved or exported as *.jpeg format (Figure 6.23).
Figure 6.22: Pop up window showing priority score and ranking detail of a road
Figure 6.23: Saving or exporting map output
Button for Saving and

Exporting Map
Page93
6.7. Data preparation and data updating for RPPM
RPPM uses both the road database and GIS database of LGED. The road database of LGED consists of 61
tables. Some of these tables are connected with other tables. Two new tables (Road fielddata and
VehicleAttribute) need to be prepared before RPPM can be run. For the purpose of this study these two
tables were created for the district of Tangail. However, the VehicleAttribute table will remain the same
for all districts until new values on VOCs and TTSs are available. Until that time this table can be simply
copied to the road database of another district. However, the Roadfielddata table has to be created for
a district before RPPM can be run for that district.
The Roadfielddata table has 12 fields (Figure 6.24). Of these 12, six fields (the six left most columns of
the figure) are identifiers for the road while for the other six data has to be collected from field. These
fields contain the data on connectivity, local priority and road safety. At the moment the table does not
need to be updated. Initially a field engineer or a responsible officer at a field office can fill out the fields
concerning connectivity (Neighbouring Connectivity, Upazila Level Connectivity, Union Level
Connectivity, Village Level Connectivity) for all roads in an upazila. These fields provide information
whether a particular road connects to a neighbouring upazila, important facilities within the upazila,
union or village. If it meets the condition then a value of 1 (one) would be given as input in the record
otherwise it would be 0(zero).
Similarly, if road safety is an issue for a road (i.e. the road is not safe) then road safety field would show
value of 1 (one) otherwise it would show 0(zero). The fields of connectivity may also be updated at the
local level workshop but more importantly the data on local priority (local priority) for development
roads are to come from local workshop. This data would come after local level workshop. The value of 3
(three) would be given for the highest local priority while a value of 1 (one) would be given for roads
that have the lowest priority. After these data are available local level workshops, the Roadfielddata
table can now be used for running RPPM.
Figure 6.24: Structure of road field database
Page94
The Vehicle Attribute table contains values of vehicle operating cost, standard speed (km/hour) and
Value of Time (Tk/hour) in different (Good, Fair, Bad) roughness condition for 15 types of vehicles that
operate on rural roads of Bangladesh. These values are used for CBA. These values may be updated
when necessary.
RPPM also uses 7 tables of the existing LGED road database (Table 6.2). These tables contain
engineering survey data, traffic survey data, maintenance information and information on socio-
economic facilities. The data on these tables need to be regularly updated.
Table 6.2: The tables of LGED road database that is used by RPPM
Data Table Contains Data On Data Field Used in RPPM
 Gc_SM_ID,
GC_SM_List Data on Growth Center and Rural Markets
 GC_SM_Tag
RoadGC_SM_Posit Data on Position of Growth Center and Rural
ion Markets on Roads
 GcOrHatID
 RdTypeCode,
 AADT_MT,
RoadInventory Data on Physical and Traffic Features of Road  AADT_NMT,
 CVD,
 IRI
Data on socio-economic facilities (except market  Socio_Sl,
RoadSocioConn
and growth center) along the road.  S_Type_ID
RoadSegments Segment wise physical feature information of road  SurfType
 Rutting_L,
 Rutting_H,
 Depression_L,
 Depression_H,
 Ravelled_L,
Ravelled_H,
Segment wise physical condition information of
RoadSegCond  Crack_L,
road
 Crack_H,
 PotHole_L,
 PotHole_H,
 EdgeDistress_L,
 EdgeDistress_H,
 E_Date
 VehiID,
RoadTrafficCount Mode wise Traffic Volume data on the road  CountHatDay,
 CountNonHatDay
Page95
The third column in Table 6.2 provides the list of data that are needed to run RPPM. These data are
regularly collected by LGED as they are part of the existing road database and may require updating
before running the program. A short description on the data contents of the road database tables that
are used by RPPM (named in the first column of Table 6.2) is presented below.
RoadInventory is one of the most important tables for running RPPM. It contains data on type of road,
total length, crest width, traffic volumes for motorised, non-motorised and commercial vehicle, road
condition, etc. RPPM uses data from type of road (Rd Type Code), traffic volume (AADT_MT, AADT_NMT,
CVD), road condition (IRI) fields. It should be noted here that the roads (which are not fully BC) do not
have IRI value. The number (Socio_SL) and type (S_Type_ID) of socio-economic facilities along the road is
stored in road SocioConn table. The table is connected with the RoadInventory table and is used for
calculating MCA score as well as generating core network.
The RoadSegments contains information on physical survey including road surface (Surftype). On the
other hand the RoadSegCond table provides the data on condition of road. It includes data from visual
inspection of road like rutting, cracks or potholes. There are twelve fields (mentioned in Table 6.2) in this
table which are used by RPPM. This table is used for earthen and brick paved roads which do not have
IRI values to make qualitative assessment of their surface conditions. GC_SM_List table contains data on
Growth Centre and Rural Markets. Data on whether it is a growth centre or rural market (described as
Hat by in bangla), its area, market day etc.
The RoadGC_SM_Position table contains data of the roads which surrounds a growth centre or rural
market. The table is connected with GC_SM_List and RoadInventory table. The software uses the road
id and Gc or Hat ID to identify the number of growth centres / rural markets on a particular road.
RoadTrafficCount table contains mode wise (VehiID) traffic data on “hat” days (Count Hat Day) and non-
hat (Count Non Hat Day) days.
6.8. Conclusion
It is understood that software development is a continuous process and there is always some scope for
improvement. However, the development team faced two issues which, if not addressed properly, may
undermine the outcome of the whole exercise. The first issue is the quality of data. The road surface
condition and traffic volume are two important parameters for priority evaluations. However, in many
cases these data are not upto date. For example, the traffic volume data on market and non-market
days were last updated in September 2014. Similarly, the road surface condition data was last updated
in May 2015. So in some cases the results might not truly reflect the reality. It is also important to
update both the data at the same time. The second issue is the incompatibility of information in the
road database and the GIS database. Although the GIS database is integrated with the final output
produced by this study, most of the spatial features in the road database do not have spatial data. As a
result of this gap in data, spatial features (with the exception of roads) could not be shown in graphical
form.
Page96
7 Discussion on outputs generated
7.1. Introduction
According to LGED database there are 2595 roads in 12 upazilas of Tangail. There are 112 upazila roads
and 118 union roads. The rest of the roads are village roads. The total length of the roads is 8249.23 km
of which 940.39 km is Upazila road. Table 7.1 provides the distribution of number of roads in different
upazilas of Tangail.
Table 7.1: Number of Roads in different upazilas of Tangail

Name of Upazila Upazila Road Union Road Village Road Village Road
(Type A) (Type B)
No Total No Total No Total No Total
Length Length Length Length
(in KM) (in KM) (in KM) (in KM)
Basail 9 60.99 16 95.90 42 133.38 103 199.79
Bhuapur 4 25.80 10 49.64 89 196.88 159 160.53
Delduar 8 64.18 11 77.33 64 217.35 78 167.49
Dhanbari 5 41.99 15 79.56 82 201.38 22 41.28
Ghatail 10 102.48 18 117.67 144 549.37 104 338.25
Gopalpur 11 73.60 11 49.56 75 179.79 167 263.91
Kalihati 17 126.52 22 124.84 89 349.01 65 210.73
Madhupur 8 52.79 14 99.94 80 331.09 88 278.95
Mirzapur 9 101.59 19 120.55 96 337.71 241 495.19
Nagarpur 10 100.36 18 95.65 119 365.7 107 270.3
Shakhipur 11 108.79 14 140.56 58 307.89 79 225.92
TanagailSadar 10 81.30 20 116.34 85 222.14 59 97.27
Total 112 940.39 188 1167.54 1023 3391.69 1272 2749.61
RPPM generates the priority lists for improvement, further improvement and maintenance of these
roads on the basis of their surface type. While earthen roads are only considered for improvement, HBB
and partially paved roads are considered for all kinds of development. On the other hands paved roads
are only considered for maintenance and further improvement. In addition, RPPM also generates core
road network for each upazila. Maps of the core road network are provided in Annex I, volume 2.
7.2. Core road network
The maps reveal that the core road network of the district is almost unbroken if one considers the RHD
network. But in some upazilas (for example Gopalpur, Bhuapur, Mirzapur) there is scope for expanding
the core network. The number of roads in core network varies from 10 (Bhuapur and Gopalpur) to 39
(Kalihati). However the number depends neither upon the size nor upon the road density of the upazila.
In the core network of Tangail district there are 77 upazila roads and 154 union roads. List of roads
forming core road network are given in Annex J, Volume 2. There is no other category of road in the core
network. But it should be borne in mind that RPPM selects only those roads in core road network for
Page97
which data is available in LGED road database. Hence, roads maintained by RHD (National and Regional
Highways and Zila Roads) are not part of core network generated by RPPM.
7.3. Improvement of roads
RPPM generates four tables for improvement of roads for the 12 upazilas of Tangail. Generated output
of roads requiring improvement are provided in Annex K1 – Annex K4, Volume 2. Each table makes the
prioritised list for a particular type of road surface.
Depending upon traffic volume, two types of earthen roads were considered for improvement. The
roads with AADT value over 200 are considered as high volume earthen road; the other earthen roads
were considered as low volume earthen roads. However it was not possible for LGED, to collect traffic
volume data of all the earthen roads. The AADT value was stored as none (0) in the LGED road database
for those roads for which traffic data were not collected. So for some roads the priority ranking for
improvement may change if traffic volume data is collected. It was found that most of the roads of
Tangail have MCA value less than 20. There are only 90 roads which have MCA value over 20. It should
be mentioned here that of these 90 roads, AADT values for 10 roads are not available. Of these 10, eight
are located in Ghatail upazilla. Basail-Ishorgonj GC via Borni kishory Road (Road id 393092010) is the
highest ranked low volume earthen road of the district. This is an upazila road which connects Ishorgonj
GC with Basail-Sunna GC road. With the exception of Delduar upazila the top three ranked low volume
earthen roads have a MCA score over 20 for all the upazila. Table 7.3 provides the top 20 low volume
earthen roads which needed to be improved in Tangail districts.
There are 79 earthen roads with over 200 AADT. For these roads both EIRR and MCA were conducted.
Deuli UP office (Silimpur)-Bangrail Road (Rd Id 393233004) at Delduar Upazila was found to be the
highest ranked earthen road with high volume of traffic in the district. The road situated in Delduar
upazila has an EIRR value of 33.94. It should be mentioned here that all the upazilla do not have high
volume traffic in earthen road. Only seven upazila (Basail, Delduar,Kalihati, Mirzapur, Nagarpur,
Shakhipur and Tanagail Sadar) have high volume earthen road. Considering an EIRR of 8.5%5, it was
found that only nine roads (Table 7.4) could be selected. If 12%6 is considered as cut off value for EIRR
only eight roads could be selected. Most of the roads have an EIRR value below 6 (Figure 7.1a). Though
some of the roads have high MCA score (there are six roads which have an MCA score over 50) most of
the roads have MCA score below 20 (Figure 7.1b). It is interesting to learn that only three top ranked
road on MCA are in the top 20 in high volume earthen roads priority ranking. It is also should be noted
here that a good number of roads have negative EIRR.
5
the lowest interest rate for term loans of financing institutes of Bangladesh in July 2017 (Bangladesh Bank, 2017)
6
the interest rate for term loans of most of the financing institutes of Bangladesh in July 2017 (Bangladesh Bank, 2017)
Page98
Figure 7.1a: Distribution of roads for improvement (high volume earthen road) according to EIRR
Figure 7.1b: Distribution of roads for improvement (high volume earthen road) according to MCA value
Page99
The partially paved roads have reasonable traffic. The average AADT of these roads is 314.703 and in a
number of roads considerable number of commercial vehicles ply. However it should be kept in mind
that LGED did not collect traffic data for all the partially paved roads; so some of the roads may be
ranked higher once data for AADT and CVD is there. It was found that Nagarpur H.Q-Daragram GC via
sanka bazaar (Road no 393762007) of Nagarpur upazila was the highest ranked partly paved road. This is
one of the longest roads of Tangail district of which around 50% is paved. It also had the highest AADT
value of all the partly paved roads of Tangail. If we consider 8.5% and 12% EIRR for investment, 120 and
56 roads could be selected for improvement. However, this criteria is strictly followed then some high
volume traffic road (for example, Habla UP Office-Karatia GC via Motora Road, Road Id393093007 of
Basail Upazila with a AADT of 1889 and CVD of 529) may not be selected while some low volume traffic
road (For example Road No 393854053, Bankichowrasta - Shalgrampur Road with AADT of 130 and CVD
of 23) may be selected. It may be mentioned here that like high traffic volume earthen road, some of
the partially paved roads generate negative EIRR value. Table 7.5 provides the list of top 20 ranked
partially paved road in Tangail. The table demonstrates that some of the upazilas are over represented
and some of the upazila (for example Tanagil Sadar, Mirzapur) have no roads in the top 20. The MCA
values for most of these roads are high. There are more than 50 partially paved roads in Tangail with
MCA value over 50 and none of the roads has MCA value of 0.
The Bonki Gramin Bank R&H - Bonki Mirza Chala Road (Road id 393855063) of Shakhipur was highest
ranked HBB road which needed improvement in Tangail. There were only nine roads in this category and
most of these have EIRR value of 0. In fact Bonki Gramin Bank R&H - Bonki Mirza Chala Road had an
EIRR below 8.5%. The roads also had low MCA score (the highest being 17.281).
7.4. Maintenance of roads
For maintenance of roads, MCA was the criteria for ranking. As mentioned earlier, earthen roads were
not considered for maintenance. Generated output of roads requiring maintenance are provided in
Annex K5 to Annex K7, Volume 2.
The existing data shows that 70 partially paved roads needed maintenance work. Compare to partially
road for improvement it had higher average MCA score (25.129 and 40.306 respectively). Figure 7.2a
provides a graphical representation of the distribution of the MCA of the roads that fulfil the criteria to
be in the maintenance list. Incidentally, the top ranked road Nagarpur H.Q-Daragram GC via sanka
bazaar (Road no 393762007) of Nagarpur upazila was the highest ranked partly paved road for
improvement also. It should be mentioned here that of the top 20 partially paved roads that needed to
be maintained on priority basis in Tangail districts (Table 7.6) nine were also listed in top 20 prioritised
roads for improvement in the district. So during decision making both the list needed to be considered
together. Kuripara Simarpar to Masidpur Road (Road no 393665188) at Mirzapur Upazila was the only
HBB road that fulfil the criteria for maintenance. On the other hand, there were 94 fully paved roads
which fulfil the criteria. As expected these roads had heavy volume of traffic (both AADT and CVD).
These roads also had very high MCA scores (Figure 7.2b for distribution of MCA value). Sakhipur-
Sagardighi via Barochowna Road (road id 393852001) at Shakhipur Upazila was the top ranked paved
road for maintenance in the district. Table 7.7 reveals that of the 20 top ranked paved roads five are
located at Kalihati upazila and four at Ghatail Upazila. On the other hand, no paved roads from
Dhanbari, Mirzapur and Tangail Sadar were in top 20 list for maintenance of paved roads
Page100
Table 7.3: Top ranked low volume earthen roads for improvement in Tangail
Ranking Road Code Road Name Upazila AADT CVD Length MCA
in km Score
1 393092010 Basail-Ishorgonj GC via Bornikishory Road. Basail 131 7 3.43 70.81
2 393763016 Nagarpur-Shahbatpur UP road niaNagarpur Govt. College Nagarpur 200 0 3.00 57.26
3 393234018 Delduar Bazar-DelduarElasin Road Delduar 134 13 0.50 41.96
4 393384055 Jamtail-JhawailFerryghat Gopalpur 9 0 2.55 40.66
5 393964072 Keramjani-HadiraU.P.office via Pankata Bazar road Dhonbari 59 0 5.00 37.73
Shahjanipucca road at Shajhani High School via Atapara Primary
6 393765034 Nagarpur 131 0 2.62 34.76
School JamunaGhat
7 393575018 Dhalpur-Malka road Capri Modhupur 40 0 3.85 33.88
8 393764078 Tangail-Aricha road at BRCK office to Shahbatpur road. Nagarpur 128 0 2.00 33.71
9 393853009 Kalia UP office-Debraj Bazar Rd Shakhipur 24 0 6.00 33.16
10 393194068 Shailkol hat-Lakhipur road Bhuapur 4 0 2.10 32.89
11 393194022 Amula-Changtapara road Bhuapur 4 0 4.00 32.89
Guiagomvir (Makrai Bazar)-Soankhola Bazar (Sandhanpur UP
12 393283007 Ghatail 0 0 5.00 31.95
office) Road
13 393665011 Jamurki-Dubail Rd Mirzapur 48 0 2.10 31.46
14 393665035 Zamurki High school to Dobail road Mirzapur 57 0 1.50 31.15
15 393764065 Khorshed Market to Bharrah GC road via Bir-salil Nagarpur 178 0 2.50 30.97
16 393764084 Chandakmadrasha to Subantalimadrasha via Agdhiguliabazaar Nagarpur 135 0 4.20 30.89
17 393854025 Kalidash-Borachala via Raja Bari road Shakhipur 77 0 10.75 30.65
18 393384056 JhawailBagdibari-Darisaya P/School Gopalpur 71 0 2.27 30.63
19 393765066 Singdair bazar-south Taluknagar road. Nagarpur 22 0 1.50 30.30
20 393194064 NiklaDoripara Based House-KhorakChanmunshi House road* Bhuapur 4 0 2.40 30.19
Page101
Table 7.4: Top Ranked high volume earthen roads for improvement in Tangail
Ranking Rd Code Rd Name Upazila AADT CVD Length in EIRR MCA
km
1 393233004 Deuli UP office (Silimpur)-Bangrail Road Delduar 1004 80 3.95 33.95 47.20
2 393473003 Bangra U.P.-Sahadebpur Bazar Road Kalihati 892 47 2.10 24.09 43.44
3 393953011 Bashakhanpur Bazar-Hugra UP via Beguntola Bazar TangailSadar 811 50 9.00 20.93 33.87
4 393234029 Babupur-Bangrail Delduar 492 70 2.45 19.34 11.66
5 393233008 Deoli Up Office-Elasin GCC Road Via Shalpabatia Delduar 740 12 3.49 16.73 44.77
Kanchanpur UP Office-Patherghata hat via Tarabari
6 393093011 Basail 434 48 8.70 14.63 28.61
Road.
7 393092009 Ishorgonj GC-Sunna GC Road. Basail 388 30 3.00 14.39 75.87
8 393663015 Warshi R&H road to Warshi UP road Mirzapur 679 70 3.00 13.17 21.84
9 393953006 Baghil-Aynapur Road TangailSadar 505 74 5.73 10.04 44.36
10 393473014 Elanga U.P.-Nardahi Bazar road Kalihati 540 11 5.00 8.97 60.87
11 393093008 Basail-Kanchanpur UP office Road. Basail 344 48 3.92 8.07 41.30
12 393234001 Putiajani-Meruaghona Rd Delduar 414 17 2.50 6.48 16.35
Mahanandapur Bazar-Baheratail UP via
13 393853015 Shakhipur 267 4 7.60 6.06 47.26
HarangaChala Rd
14 393093014 Fulkihat(Janjania)-Kawaljani UP via Badiajan. Basail 238 10 4.00 5.76 28.74
15 393234026 Deoli hat to jhunkai hat Road Delduar 293 14 1.00 4.91 8.89
16 393473015 Kokdahara U.P.-Balla U.P. Road Kalihati 547 0 3.00 4.77 41.00
17 393472014 Jamuna Bridge-Torabgonj G.C. road Kalihati 507 21 3.70 4.51 57.81
18 393094024 Karatia-Saysata via Sonalia Road. Basail 202 19 2.70 2.76 5.54
19 393234039 Natiapara-Basail Road Delduar 221 28 0.70 2.70 16.81
20 393473017 Shahadebpur U.P.(Pouzan)-Bhukta bazar Road Kalihati 434 17 4.25 2.61 33.22
Page102
Table 7.5: Top ranked partially paved roads for improvement in Tangail
Ran- Road Code Road Name Upazila AADT CVD Length EIRR MCA
king in KM
1 393762007 Nagarpur H.Q-Daragram GC via sanka bazar Nagarpur 2348 45 10.55 73.55 85.97
2 393852004 Sakhipur - Suruj GC Road via Salgrampur,TejpurFerryghat. Shakhipur 1666 186 11.60 71.27 79.68
3 393853014 Shakhipur (Banki)-TatuliarChala via Futanir Bazar road Shakhipur 1467 292 4.20 68.62 56.19
4 393472002 Bhandeshwar G.C-Balla G.C. Road Kalihati 843 200 4.65 52.73 77.61
Moghalpara Bazar-Singuria Bazar via Dighulkandi, Anehola
5 393283001 Ghatail 1258 304
UP Office Rd. 12.37 50.66 55.72
6 393853010 Berbari Bazar-Boheratoil UP via NakshalaBetua Shakhipur 942 111 10.15 46.50 39.85
Basail (Andirapara)-Habla UP Office via BalinaVoirpara
7 393093016 Basail 1130 233
Road. 8.10 45.23 48.59
8 393762003 Nagarpur-Mirzapur Road Via Mokna Nagarpur 1352 190 15.78 44.51 78.98
9 393282004 Porabari-Angarkhola-Garo Bazar Rd. Ghatail 1244 158 21.25 41.96 82.89
10 393474060 Kalihati(Haripur) - Nagbari U.P. Road via Bhandeshwar G.C. Kalihati 813 29 6.00 40.70 33.63
11 393853002 Hatibanda UP (Taktarchala)-Bazail Bazar Shakhipur 813 90 9.45 37.25 59.57
12 393853001 Jadabpur UP (Nalua)-salda Bazar road via beltoil Shakhipur 694 121 14.01 34.49 60.75
Birtibari R&H (ModhupurUpazila) - Dhanbari GC via
13 393572011 Madhupur 618 149
Zagirachala Road 9.07 32.81 40.95
14 393232006 Tangail-Aricha R&H Road to Lowhati GCC Road. Delduar 832 113 9.87 32.31 73.39
15 393093005 Kashil UP Office-Bilpara hat via Kashil High School. Basail 1024 137 5.50 31.84 44.02
16 393854027 Kalidash-Beltoil road Shakhipur 399 127 4.00 30.40 17.52
17 393093001 Basail-Kawaljani UP via Maizkhara. Basail 901 52 7.51 30.39 53.15
Baoikhula Bazar-Nallapara hat via Dubail UP office, Bathuli
18 393233002 Delduar 1113 139
hat, Fultara Hat Road 16.13 29.74 48.15
Shakhipur (Mojib College) to Solanga via Isadighi Bazar
19 393853006 Shakhipur 617 68
PatherChowrasta 17.05 29.34 52.03
20 393093009 Kashil UP-Nayerhat via Balia Road. Basail 607 224 6.66 29.27 36.55
Page103
Figure 7.2a: Distribution of MCA value for partially paved road selected for maintenance
Figure 7.2b: Distribution of MCA value for paved road selected for maintenance
Page104
Table 7.6: Top ranked partially paved roads for maintenance in Tangail
in KM
1 393762007 Nagarpur H.Q-Daragram GC via sanka bazaar* Nagarpur 2348 45 10.55 88.47
2 393662005 DeohataDhantara G C road Mirzapur 829 167 8.80 86.52
3 393092004 Basail-Natiapara GC via Bilpara Road. Basail 648 117 9.66 84.53
4 393472002 Bhandeshwar G.C-Balla G.C. Road* Kalihati 843 200 4.65 80.90
5 393232006 Tangail-Aricha R&H Road to Lowhati GCC Road* Delduar 832 113 9.87 80.43
6 393092005 Basail-Sunna GC Road. Basail 720 137 5.68 80.42
7 393472001 Kalihati (Dhunail)-Shayahat-Hatia JBA Road Kalihati 813 48 10.00 77.02
8 393762006 Nagarpur HQ-Louhati GC Via Khamardhalla Road Nagarpur 767 68 8.70 71.77
9 393853002 Hatibanda UP (Taktarchala)-Bazail Bazar* Shakhipur 813 90 9.45 66.57
10 393093002 Kashil UP Office-Aisara Bazar via Janjania. Basail 906 100 11.86 64.51
11 393093001 Basail-Kawaljani UP via Maizkhara.* Basail 901 52 7.51 63.65
12 393853003 Hatibanda UP (Taktarchala)-Koratali Ferry Ghat via 690 53 7.25 62.96
KamaliaChala Bazar Shakhipur
13 393283001 Moghalpara Bazar-Singuria Bazar via Dighulkandi, Anehola UP 1258 304 12.37 62.32
Office Rd*. Shakhipur
14 393853014 Shakhipur (Banki)-TatuliarChala via Futanir Bazar road* Shakhipur 1467 292 4.20 61.88
15 393473007 KalihatiUpazila-Narandia U.P. Road Kalihati 541 26 9.90 55.24
16 393572011 Birtibari R&H (ModhupurUpazila) - Dhanbari GC via 618 149 9.07 54.82
Zagirachala Road* Madhupur
17 393473006 Rampur Bhashani G.C.-Fulki U.P. road Kalihati 588 39 3.30 53.15
18 393473020 Nagbari U.P. - Kawalzani U.P. Road via Ratangonj Bazar Kalihati 1068 65 7.20 52.77
19 393473005 Paikara U.P(Singutia)-Pouli Bazar Road via Mohela Kalihati 629 22 5.70 51.64
20 393093005 Kashil UP Office-Bilpara hat via Kashil High School* Basail 1024 137 5.50 51.39
Note: Road named marked with * is also in the prioritised list of top 20 partially paved roads for improvement
Page105
Table 7.7: Top ranked paved roads for maintenance in Tangail
in KM
1 393852001 Sakhipur-Sagardighi via Barochowna Road Shakhipur 1675 561 14.66 95.74
2 393282002 Sakhipur-Garo Bazar Road Ghatail 2188 900 15.26 92.67
3 393472010 Kalihati RHW-Barachowna G.C. Road Kalihati 2396 660 10.42 90.81
4 393282003 Hamidpur-Deopara Road Ghatail 1643 478 11.25 88.59
5 393852009 Barachowna-Kalihati via Chakpara Shakhipur 1385 369 11.00 88.47
6 393092001 Basail-Kanchanpur GC Road. Basail 582 79 3.70 87.97
7 393852002 Sakhipur-Kalidash Road Shakhipur 779 193 5.74 87.17
8 393282001 Ghatail-Jamuria-Bhuapur Road Ghatail 1486 307 11.33 86.07
9 393382008 Darisoya-Bhenglabazaar Gopalpur 380 46 3.25 84.67
10 393282010 Makrai-Deopara Road Ghatail 1275 193 7.65 83.72
11 393472016 Balla G.C.-Baharatail GC via Ratangonj Bazar Road Kalihati 1042 75 3.15 82.80
12 393762005 Lauhati GC-Saturia HQ Road Nagarpur 1742 250 13.00 82.16
13 393232002 Lauhati-Pakulla Rd Delduar 1732 212 3.62 80.85
14 393382002 Gopalpur-Bhengula Gopalpur 816 172 4.00 79.20
15 393472009 Zokerchar JBA-Nikrail G.C. Road Kalihati 1468 77 2.00 78.77
16 393472013 Shaya G.C.-Shialkol G.C. Road Kalihati 577 28 2.71 78.20
17 393473001 Narandia-Salla Road Kalihati 1066 33 8.00 77.20
18 393572005 Raktipara-Chapri Road MadhupurP 375 72 7.50 74.94
19 393382010 Bhengula-Pingna Gopalpur 330 20 1.75 74.80
20 393762004 TangailAricha R&H at Khorshed Market-Bharra Bazar 545 46 3.55 73.32
Via Arrah Nagarpur
Page106
7.5. Further improvements of roads
Sometime it becomes necessary to improve the road geometric standards, to raise the embankment and
to widen pavement and/or road crest and to raise road embankments of an existing road. The road
surface may be fully paved, partially paved or HBB. RPPM generates a single list for all kind of road
surface for further improvement. Generated output of roads requiring further improvement are
provided in Annex K8, Volume 2. From the data, it was found that only 19 roads need further
improvement at the moment (Table 7.8). In addition at the moment not all the upazilas have roads
which need further improvement. It was found that though only seven roads had an EIRR over 8.5% or
12%, almost all the roads have MCA value over 60. It should be also mention here that all most all the
roads except were upazila road.
7.6. Conclusion
The chapter reveals that though there are many roads for improvement, further improvement and
maintenance the lists generated by RPPM should not be viewed in isolation. It is also necessary to look
from district perspective as there may be cases where deserving roads may be prioritised low in one
upazila with higher EIRR and MCA value than roads of another upazila which may be prioritised high
with lower EIRR and MCA value. One of the issues need to be addressed is the issue of authenticity and
reliability of data. This is specially true for those roads for which IRI data was not available and research
team have to rely on data on visual inspection. This may be the reason why application of RPPM resulted
only a single HBB road selected for maintenance. Similarly the research team also felt that without
traffic data for all the roads, some roads may get higher priority (as these have AADT data) ranking than
others (which do not have AADT data).
Page107
Table 7.8: Ranking of Roads considered for Further improvement in Tangail
Ranking RoadCode RdName Upazila Type of AADT CVD L

Road i
1 393662008 GoraiShakhipur RHD(Start from 1903 321
Upazila
PakuaBusstand) - Kanchanpur GC via Mirzapur
Road
Patharghata bazar road.
2 393952004 Karatia G.C. (Start from RHD road)- TangailSa Upazila 975 437
SilimpurG.C.Road. dar Road
3 393663003 Dubail RHD (GPS) - Fatepur UPC Road Via Upazila 1788 385
Mirzapur
Mohera UPC & PTC. Road
4 393472011 Upazila 1562 339
Suruz G.C.-Dhalapara G.C. Road Kalihati
Road
5 393953001 Tangail (Main Road) - Kakua Bazar via TangailSa Union 1140 361
DhalanSibpur Bazar, Kakua UP Road. dar Road
6 393282008 Upazila 1262 401
Dhalapara ( R & H ) - Chapri G C Road Ghatail
Road
7 393662002 Mirzapur HQ RHD - NagarpurUpazila HQ Union 1619 301
Mirzapur
Via Barinda Hat, Kedarpurbazaar Road. Road
8 393662001 Pakulla RHD - Lawhati GC Road Via Upazila 2424 587
Mirzapur
Bhabkhanda Bazar. Road
9 393282002 Upazila 1648 653
Sakhipur - Garo Bazar ( G C ) Road Ghatail
Road
10 393282001 Ghatail H/Q - Jamuria - Patchtikri R & H Upazila 1500 195
Ghatail
More Road Road
11 393472005 Upazila 5700 532
Elenga G.C.-Balla G.C Road Kalihati
Road
12 393472010 Upazila 1527
Kalihati RHD-Barachowna G.C. Road Kalihati
Road
13 393762005 Upazila 1264 93
Lauhati GC--Saturia UZHQ Road Nagarpur
Road
14 393952003 Tangail (Tangail-NagarpurReg.HW At 2890 317
TangailSa Upazila
Kagmari College)-Charabari-Torapgonj
dar Road
G.C.- Shahjani G.C.Road.
15 393232005 Upazila 1952 344
Silimpur GC - Karatia GC Road. Delduar
Road
16 393232001 Upazila 3708 586
DelduarUpazila H/Q - Lowhati GC Road. Delduar
Road
17 393232002 Upazila 2529 337
Lowhati GC - Pakulla R&H Road. Delduar
Road
18 393852001 Upazila 1917 259
Sakhipur-Sagardighi via Barochowna Road Shakhipur
Road
19 393962001 Upazila 1044 500
Dhanbari-Kendua Road Dhonbari
Road
Page108
8 Summary, conclusions and recommendations
Rural roads have a vital role in rural development in developing countries. Often, rural road
development is considered as an entry point to rural poverty alleviation. As found in many
evaluation studies on rural roads in Bangladesh, the development of rural roads has contributed
immensely to the social and economic development of rural people and transformation of the rural
economy in Bangladesh.
LGED has already developed a vast network of rural roads in Bangladesh including about 74,000 km
of paved roads and about 20,000 km of brick paved roads which may be paved in near future. It is
important that the selection of roads from a large number of roads is undertaken through a rational
selection process. This is necessary for the effective utilisation of available resources. Any further
development of roads should be selective and based on some criteria so that the network can be
sustained and properly maintained over a longer term. Therefore, there was a need to develop a
planning and prioritisation methodology for the development, upgradation and maintenance of rural
roads.
The project team in collaboration with LGED officials has developed a practical methodology to fulfil
this need of future rural road development through a rational selection process. The methodology
considers a set ofsocial, economic and access and connectivity related factors to prioritise three
types of road development work undertaken by LGED. The methodology was successfully
implemented in the pilot district of Tangail.
The project team has also developed and implemented a web-based road planning and prioritisation
software RPPM to replicate the methodology by LGED in other districts of Bangladesh. However,
before the running software for another district, local workshops are required to be organised to
collect some additional data. The database for that district has to be prepared with this additional
data and updated, as needed.
In developing the methodology the project team had faced a number of challenges. Some
observations, suggestions, recommendations and conclusions are outlined in this chapter for their
consideration by LGED.
Currently LGED maintains two databases namely the road database and the GIS database. These two
databases are not fully integrated. There is a wealth of information in these two databases which
could be effectively used in undertaking many useful interactive analyses relevant to rural road
development and other development purposes. As such, an integrated database system has the
potential of becoming a valuable resource for LGED as well as other government departments, for
example, government departments related to primary and mass education, secondary education
and health service. Once an integrated database is available, many useful analyses are possible even
from within the GIS platform without requiring a custom-made program. Although it may require
considerable amount of time, resources and efforts to integrate the two databases, it would be
worth considering given its potential use.
The limitations in displaying all features of the outputs produced by the planning and prioritisation
software on a map are due to this problem of the two databases. Once this integration problem can
be resolved, these limitations can be overcome and interesting and interactive displays of outputs of
interest to policy-makers can be produced.
Page109
The quality of output of any software depends on the quality of input data. Before replicating the
methodology in other districts, it is advisable that the road database, especially data on traffic
volume, be updated. LGED may consider updating the database on a regular basis, say every three
years. Both the CBA and MCA analyses use traffic data. While developing the software, the project
team has noticed that the CBA results, as expected, are sensitive to traffic volume data.
There are some 400-600 rural roads in a typical upazila. It would not be practical to undertake
volume studies for long hours on all roads. LGED may consider to develop a practical methodology to
estimate traffic volume and composition on each type of road based on traffic counts for limited
hours on representative roads for each type of road and their surface conditions. As village roads are
mostly used by pedestrians, it would be very useful if pedestrian counts, especially on village roads,
can also be included in the survey.
The hourly variation of traffic flow on upazila and union roads, many of which may require further
improvement in near future, also needs to be established. This would be required to determine if
the traffic flow on a road has already exceeded or in near future may exceed its estimated capacity.
The future projection of traffic is very important for CBA. While the project has considered the
results of available studies on traffic growth and used rational rates for the growth of normal and
generated traffic, any details growth rates by road type were not available. It is unlikely that traffic
grows at similar rates for all types of roads and for all types of vehicles. Traffic growth may also vary
by localities and other factors. The project team has observed that the results of CBA are very
sensitive to traffic projection. As such, reliable growth rates by road and vehicle type would be very
useful.
LGED may consider to address this problem to make CBA results more reliable. No major effort may
be needed to address this problem. It is understood that only a limited number of roads are
considered for improvement in a year. The Upazila Engineer can undertake volume counts for before
and after situations for these roads. Ex-post project traffic counts may be done for 5 consecutive
years. Traffic counts will also be necessary on some other selective roads outside the influence area
of the project road to establish the normal traffic growth in that area. The availability of such traffic
count data would help to establish reliable traffic growth rates by type of road as well as by district.
As expected, CBA for low volume village roads did not produce any meaningful results. In fact, for
most of the low volume roads it was not possible to estimate the very low (negative) EIRR values
because of computational issues. The village roads are mostly used by pedestrians. If pedestrian
traffic counts are available, pedestrian traffic volume may be included in MCAs as a criterion. The
results of MCAs then can better reflect the actual use of low volume village roads. LGED may
consider to include pedestrian traffic countsin the road database, especially for village roads. The
availability of pedestrian traffic data and its inclusion in MCA can make the results of such analyses,
especially priority ranking, more robust and representative.
Page110
References
Ahmed, R. and Hossain, M. 1990. Development Impact of Rural Infrastructure in Bangladesh. Research Report
83, International Food Policy Research Institute (IFPRI) in collaboration with Bangladesh Institute of
Development Studies (BIDS).
Airey, A. 2014. Good Policies and Practices on Rural Transport in Africa: Monitoring and Evaluation. Working
Paper no. 99, Sub-Saharan Africa Transport Policy Program.
Archondo-Callao, R. 2004. The Roads Economic Decision Model (RED) for the Economic Evaluation of Low
Volume Roads - Software User Guide & Case Studies, SSATP Working Paper No. 78, World Bank.
Archondo-Callao, R. 1999. Roads Economic Decision Model (RED) for economic evaluation of low volume
roads. Sub-Saharan Africa Transport Policy Program (SSATP) technical note; no. 18. Washington, DC:
World Bank.
Asian Institute of Transport Development. 2011. Socio-economic Impact of National Highway on Rural
Population, New Deli, India.
Bangladesh Bank. 2017. Announced interest rate chart of the scheduled banks (lending rate)
(Percentage per annum), Available at: https://www.bb.org.bd/fnansys/interestlending.php accessed on
30 Aug, 2017
BBS. 2015.Zila Report: Tangail, Bangladesh Population and Housing Census 2011, Bangladesh Bureau of
Statistics, Statistics and Informatics Division, Ministry of Planning, Government of the People’s Republic
of Bangladesh.
BBS. 2013. District Statistics 2011: Tangail, Bangladesh Bureau of Statistics, Statistics and Informatics Division,
Ministry of Planning, Government of the People’s Republic of Bangladesh.
BBS. 2012. 2012 Statistical Yearbook of Bangladesh, Bangladesh Bureau of Statistics, Statistics and Informatics
Division, Ministry of Planning, Government of the People’s Republic of Bangladesh.
Benmaamar, M. 2003. A Method for the Appraisal of Low Volume Roads in Tanzania. Paper for the XXII PIARC
World Road Congress, Durban
Bhandari, S.B., Shahi,P.B. and Shrestha, R.N. 2014.Multi-criteria evaluation for ranking of rural road projects:
Case study of Nepal, IOSR Journal of Mechanical and Civil Engineering, vol 11, issue 6, pp. 53-65.
Donnges, C. 2001. Rural transport and local government units: how to improve rural transport for the rural
poor?, Transport and Communications Bulletin for Asia and Pacific, No. 71, pp. 19-27, available from:
www.unescap.org/ttdw/Publications/TPTS_pubs/TxBulletin_71/bulletin71_c.pdf.
ESCAP. 1979a. Feeder Road Appraisal, in Report on Seminar cum Study Tour on Economic and Technical
Aspects of the Construction of Local Roads and Rural Transport, United Nations, Bangkok, pp. 285 –
294.
ESCAP. 1979b. Guidelines for Rural Centre Planning, United Nations, New York.
GIZ. 2014. Improving Rural Transport Infrastructure: Experience from Bangladesh – An overview on the Rural
Infrastructure Improvement Project (RIIP -1).
GTZ. 2009. Road User Cost Study For LGED Roads, Final Report, Rural Infrastructure Improvement Project (RIIP)
RDP-25, Institutional Support and Training Component, Ministry of Local Government, Rural
Development and Cooperatives, Government of Bangladesh.
GoB. 2005. Road Design Standards - Rural Road. Dhaka.
Government of India.2008. Eleventh Five Year Plan (2007-2012), vol. III, Planning Commission,
Chapter 9, accessed from http://planningcommission.nic.in/plans/planrel/fiveyr/welcome.html.
Greenstein, J. and Bonjack, H. 1983. Socio-economic evaluation and upgrading of rural roads in agricultural
areas of Ecuador. Proceedings of the 3rd Low Volume Roads Conference. Transportation Research
Record 898, pp 88-94.
Government of Sri Lanka, Annual Report 2008, Ministry of Finance and Planning.
Guhnemann, A., Laird, J. J. and Pearman, A.D. 2012. "Combining cost-benefit and multi-criteria analysis to
prioritise a national road infrastructure programme", Transport Policy, vol 23, no. 1, pp. 15-24.
Page111
Haggett, P., and Gunawardena, K. A. 1964. Determination of Population Thresholds for Settlement Functions
by the Reed-Muench Method. The Professional Geographer, 6-9.
Hiep, V. D. 2006. HDM-4,dTIMS-CT, &HERST Tools for Asset Management, Saitama University, Available at
http://www.academia.edu/1734546/Comparison_between_HDM-4_dTIMS-CT_and_HERS-
ST_Tools_for_Asset_Management accessed on 29th April 2016
Hine, J., Simon, E., Done, S. and David, K. 2003. Ghana Feeder Road Prioritisation.
Hongye F. 2012. “Rural road maintenance in China”, Transport and Communications Bulletin for Asia and the
Pacific, No. 81, Planning for Accessibility and Rural Roads. Available at:
http://www.unescap.org/sites/default/files/bulletin81_Article-4.pdf.
Kharate, V. B. 2008.Hierarchical Patterns of Rural Central places in the Painganga Valley, International
Research Journal, vol II(5), pp. 471-474.
International Development Association, 2008. “Transport: Improving services for the poor” (Washington, D.C.,
World Bank)
Khan, S. and Ahmad, A. 2013. Hierarchy of Rural Settlements: A Strategy for Balanced Regional Development in
Aligarh District, International Journal of Scientific Research, vol. 2(9), pp. 194-197.
Kumar, A. and Kumar,P. 1999. User friendly model for planning rural road, Transportation Research Record,
vol. 1652, pp. 31-39.
Lebo, J. and Schelling, D. 2001. Design and Appraisal of Rural Transport Infrastructure, Ensuring Basic Access
for the Rural Communities, Technical Paper No 496, World Bank, Washington, USA.
LGED. 2016. Road Database, Local Government Engineering Department, Local Government Division, Ministry
of Local Government, Rural Development and Cooperatives. Website: http://www.lged.gov.bd/
LGED. 2010. Planning Guidelines for Rural Road Master Plan (Guideline on GIS Application for Rural Road
Development.
LGED. 2009X. Results of Terminal Surveys on Benefit Monitoring and Evaluation, Final Report, Volume 1, Rural
Infrastructure Improvement Project (RIIP), Rural Development Project (RDP)-25, August 2009.
LGED. 2004. Feasibility Study on Sylhet Division Rural Infrastructure Development Project (SDRIDP): Draft Final
Report, Government of Bangladesh.
LGED. 2002. Regional Road Network Study for Prioritisation under RRMIMP-II, Final Report, prepared for LGED
by DHV Consultants, The Netherlands, DEVCONsultants Limited, SARM Associates Ltd. and Design
Consultants Ltd.
LGRDC. 2013. Guidelines for Rural Roads and Culverts Maintenance, Local Government Division, Ministry of
Local Government, Rural Development and Cooperatives, Government of Bangladesh.
Lombard, P. and Coetzer,L. 2007.The estimation of the impact of rural road investments onsocio-economic
development.
Oudheusden, D. L. and Khan, L. R. 1987. Planning and development of rural road networks in developing
countries, European Journal of Operational Research, vol. 32-3, pp. 353–362.
Philippine Rural Development Project.2015. The I-BUILD Operations Manual.
Planning Commission. 2015. The Seventh Five Year Plan: FY2016-FY2020, Government of Bangladesh.
Planning Commission. 2005. Project Appraisal Framework: Road Sector Manual, Planning Commission,
Government of Bangladesh.
Rahman, K. U. and Noor, S. 2005. Central Functions and Centrality in Rural Settlements of Sargodha District,
Pakistan Geographical Review, vol. 60(2), pp. 70-79.
Sarker, A. K. and Dash, M. 2011. Quantifications of accessibility and prioritisation of Villages for local level
planning, Transport and Communications Bulletin, No. 81, pp.1-22.
Singh, A. K. 2010. GIS Based Rural Road Network Planning for Developing Countries, Journal of Transportation
Engineering, Available at: http://dx.doi.org/10.1061/(ASCE)TE.1943-5436.0000212.
Shrestha, C. B. 2003. Developing a computer-aided methodology for district road network planning and
prioritisation in Nepal, Transportation Research Board, vol. 3, pp. 157-174.
Page112
Shrestha, J.K., Benta, A. andLopes, R.B. and Lopes, N. 2013. A Methodology for Definition of Road Networks in
Rural Areas of Nepal, International Journal of Civil, Environmental, Structural, Construction and
Architectural Engineering, vol. 7, No. 6, pp. 422-426.
Squire, L. and Van der Tak, H.G. 1975. Economic analysis of projects. London, The Johns Hopkins University
Press.
UNCHS. 1985. Guidelines for the Planning of Rural Settlements and Infrastructure: Road Networks. Nairobi,
Kenya.
World Bank. 2008. HDM 4 – Introduction (Powerpoint Slides) Available at
http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTTRANSPORT/EXTROADSHIGHWAYS/0,,conte
ntMDK:22011461~pagePK:148956~piPK:216618~theSitePK:338661,00.html#downloads accessed on
29th April 2016
World Bank. 2003. Toolkit for the Economic Evaluation of World Bank Transport Projects. World Bank,
Washington D.C., USA.
Yedla, S. and Shrestha, R.M. 2003. Multi-criteria approach for the selection of alternative options for
environmentally sustainable transport system in Delhi. Transportation Research Part A: Policy and
Practice, 37(8), pp.717–729.
Page113
Scalable logos for use for ReCAP, AfCAP and AcCAP reports
Page114
Page115

Prioritization of Roads - ReCAP - BUET - Draft Final Report - V1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Prioritization of Roads - ReCAP - BUET - Draft Final Report - V1

Uploaded by

Copyright:

Available Formats

Planning and Prioritisation of Rural

Department of Urban and Regional Planning (DURP)

Bangladesh University of Engineering and Technology

Quality assurance and review table

ReCAP Project Management Unit

RESEACH FOR COMMUNITY ACCESS PARTNERSHIP (ReCAP)

ReCAP is a research programme, funded by UK Aid, with the aim of

Acronyms, Units and Currencies

AADT Annual Average Daily Traffic

7 Discussion on outputs generated……………………………………………………………………… 97

8 Summary, conclusions and recommendations………………………………………………………..109

The main objectives of the project, among other things, include:

In developing the methodology the following matters have been considered:

MCA criteria Improvement Improvement of Further Maintenance of

The planning and prioritisation software

Priority list Type of road development Type Surface type

RPPM can also generate three other types of output:

Table 1.1: Types of Road and their responsible authority

Table 1.2: Status of rural Roads in Bangladesh (as of June 2016)

The Five Year Plan also identifies major strategies as follows:

 Rural infrastructure development/ improvement will be planned and implemented based on

1.2. Objectives of the research study

1.3. Scope of work

1.4. Implementation arrangements and quality assurance

1.5. Deliverables and milestones

The following are the main deliverables of the project:

Draft Report for presentation 30 August 2017 Present Report

Production of Prototype 30 October 2017 Field work in

2.2. Broad categories of planning and prioritisation methodologies

1) Multi Criteria Analysis (MCA) (also known as Prioritisation Ranking or PR techniques)

2.2.1. Multi Criteria Analysis (MCA)

2.2.2. Economic analysis (Cost Benefit Analysis)

2.2.3. Participatory approaches

2.2.4. Network planning models based on optimisation techniques

2.2.5. Hybrid methodologies – Combination of multiple methodologies

2.3. Prioritisation of rural roads in Bangladesh: Current practice

Indicator/Group Description Rating*

Classification Upazila Road 12

Surface Type Fully BC 0

Gaps Major Gaps 3

Traffic Volume AADT 1000+ 100

Hospitals Upazila Health Complex 9

Union Parishad Office 12

Educational Secondary School 6

2.3.1. Merits and demerits of the current practices

2.4. Rural road development programme in some Asian countries

The Philippines is implementing an Intensified Building-Up of Infrastructure and Logistics for

 Number of beneficiaries – the bigger the number, higher the priority.

2.5. Observations and conclusions related to developing a methodology for the

In developing the methodology the following matters were considered:

Source: LGED GIS Database, 2010

Planning and Prioritisation of Rural Roads

Review of current policies, practices and resources

The prioritisation component of the methodology considersphysical (including connectivity),

 All RHD Roads (the given roads)

3.4.2. Connectivity and local priority for development

Table 3.1: Level of connectivity of roads and their assigned scores

3.5. Prioritisation of rural roads development

The prioritisation methodologies on improvement, further improvement/upgradation and

3.5.1. Prioritisation for “improvement” of rural roads

3.5.3. Prioritisation for “maintenance” of rural roads

3.6. Multi-Criteria Analysis

3.6.1. Selection of evaluation criteria for Multi-Criteria Analysis (MCA)