International Journal of Production Research

ISSN: 0020-7543 (Print) 1366-588X (Online) Journal homepage: http://www.tandfonline.com/loi/tprs20

Determinants of scale efficiency in the Brazilian

3PL industry: a 10-year analysis

Peter F. Wanke

To cite this article: Peter F. Wanke (2012) Determinants of scale efficiency in the Brazilian 3PL
industry: a 10-year analysis, International Journal of Production Research, 50:9, 2423-2438,
DOI: 10.1080/00207543.2011.581005

To link to this article: http://dx.doi.org/10.1080/00207543.2011.581005

Published online: 09 Sep 2011.

Submit your article to this journal

Article views: 310

View related articles

Citing articles: 1 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=tprs20

Download by: [Universite Laval] Date: 17 September 2015, At: 18:27

 International Journal of Production Research
Vol. 50, No. 9, 1 May 2012, 2423–2438

Determinants of scale efficiency in the Brazilian 3PL industry: a 10-year analysis

Peter F. Wanke*

Centre for Logistics Studies, COPPEAD Graduate School of Business,

Federal University of Rio de Janeiro, Rio de Janeiro, Brazil
(Final version received April 2011)

This paper focuses on the Brazilian third-party logistics (3PL) sector which, facing increasing competition,
offers clients a wide variety of services/IT technologies in the quest to bolster efficiency. The main research
objective is to determine which variables significantly impact the 3PLs’ scale efficiency by applying a two-stage
DEA model. Based on an unbalanced panel model, secondary data from the annual study published by
Revista Tecnologı´stica (years 2001–2010) were collected and analysed. The required inputs and outputs for the
Downloaded by [Universite Laval] at 18:27 17 September 2015

analysis were identified, as were the contextual variables that can affect the scale efficiency of the 3PLs. The
results support findings in the literature regarding the role coordination processes play in logistics
performance. Managerial implications for both shippers and 3PLs are discussed.
Keywords: longitudinal study; 3PL; DEA; scale efficiency; Brazil; Tobit regression

1. Introduction
There has been a trend since the 1980s to outsource non-core logistics activities (Arroyo et al. 2006). In fact,
companies have been turning increasingly to third-party logistics providers (3PL) to carry out their logistics
activities, both in the US (Lieb and Bentz 2005, Lieb 2008) and in Europe (Carbone and Stone 2005). The logistics
services market has been growing steadily in different countries worldwide, such as Mexico and China (Arroyo et al.
2006, Zhou et al. 2008).
In Brazil, Plano Real – an economic plan launched in 1994 – and the economic stability of the country have
caused an intensification of the use of outsourced logistics functions by Brazilian shippers (Fleury and Ribeiro 2003,
Wanke et al. 2007). This occurred with a certain delay regarding other contexts, such as those in the US and in
Europe. Currently, two-thirds of the logistics-related expenditure of Brazilian firms are earmarked for logistics
service providers, a fact that underscores the importance of outsourcing for the country (CEL 2009). As such, 3PLs
must continually be on the lookout for new ways to stay competitive (Panayides 2007), with efficiency evaluation
techniques serving a fundamental role in this quest.
A powerful tool for measuring efficiency is Data Envelopment Analysis (DEA), developed over 30 years ago
(Cook and Seiford 2009). Its main characteristic is the capacity to simultaneously process multiple inputs and
outputs, thereby aiding managers in decision-making. Despite its major shortcomings, the non-parametric DEA
frontier model remains widely used in transportation/logistics efficiency research in general, probably because it has
been successfully applied to a wide number of different planning situations (see, for example, Ross and Droge
(2004), Hamdan and Rogers (2007), Lin and Tseng (2007), Zhou et al. (2008), Min and Joo (2009), and Panayides
et al. (2009)). In conjunction with multivariate data analysis techniques, DEA enables the impact of contextual
variables on efficiency scores to be measured by means of two-stage models (Cooper et al. 2007). Typically, DEA is
used in stage one to calculate efficiency scores, while stage two incorporates DEA results in the form of dependent
variables into the corresponding regression (Cooper et al. 2004, Yang 2006).
The Brazilian 3PL sector is the focus of this paper. Its objective is to identify the chief determinants of scale
efficiency from 2001 to 2010. To this end, a review of the literature was carried out, both to characterise the sector,
and to support the two-stage model adopted. More precisely, an estimation of the DEA efficiency in stage one was
followed by a Tobit regression analysis using unbalanced panel data in stage two, thus allowing the effect of
contextual variables on sector scale efficiency to be estimated.

*Email: peter@coppead.ufrj.br

ISSN 0020–7543 print/ISSN 1366–588X online

ß 2012 Taylor & Francis
http://dx.doi.org/10.1080/00207543.2011.581005
http://www.tandfonline.com
 2424 P.F. Wanke

The remainder of the article comprises six sections. Section 2 discusses the role of 3PLs in supply chains, the
main services provided and the information technologies available to be adopted. Also presented are the scant
previous studies that applied DEA to the 3PL sector in other countries. Section 3 supports the most fundamental
methodological issues concerning data analysis, not only providing a background on DEA but also giving a more
detailed presentation of the two-stage DEA model used in this research. In Section 4, the data are analysed and the
results discussed altogether with several relevant issues, such as the focus on scale efficiency, the variable selection,
the possibility of variable reduction, and the validity of using secondary data sources. Section 5 addresses
managerial implications. Finally, Section 6 presents the conclusions of the study.

2. Literature review
The definition of 3PLs, the reasons for outsourcing (Knemayer and Murphy 2004), the establishment of frameworks
for the logistics outsourcing decision (Yellepeddi and Rajagopalan 2005), the selection criteria (Ganesh 2005, Perçin
2009), and the scope of the activities 3PLs provide (Zhou et al. 2008) have been discussed in numerous publications
related to 3PLs.
Generally speaking, a 3PL is ‘an integrated logistics services provider that is prepared to satisfy all or almost all
Downloaded by [Universite Laval] at 18:27 17 September 2015

of a client’s logistics needs in a customized way’ (Fleury 2000). Cost reduction, improvement of service levels,
increased operational flexibility, and enhanced ability to focus on the core business have been reasons to hire 3PLs
(Wilding and Juriado 2004, Rafiq and Jafaar 2007, Lieb 2008). Being more specific, a search on Proquest database
for scholarly journals over the past three years reveals a number of possible drivers for the outsourcing of logistics
activities. The results of this search are presented in Table 1. As shown in Table 1, there seems to be a consensus that
the key drivers of the outsourcing of logistics activities by shippers are efficiency-related, such as cost/investment
reduction, focus on core activities, and service improvement. Other relevant drivers encompass increased control of
logistics activities, greater efficiency in execution, know-how for new logistics solutions, enhanced use of
information technology, and market expansion.
The emergence of 3PLs capable of performing a range of tasks with different levels of specialisation is a natural
consequence of the variety of previously mentioned reasons for outsourcing parts of logistics operations (Figueiredo
and Mora 2009). In a survey of large manufacturing firm users of 3PL services, conducted in Brazil, Wanke et al.
(2007) identified a strong association between the production process structure of these firms and the type of
services/technological underpinnings offered by the 3PLs. More specifically, it was evident that firms in the
automotive, electric appliance, and aerospace sector tend to hire integrated 3PLs, i.e. that handle transport, storage,
and inventory concurrently, and that depend heavily on information technologies. In contrast, foodstuff, beverage,
and fuel firms, for example, tend to hire 3PLs with less of a technology-intense approach – firms more geared
towards providing basic transportation services.
According to Lewis and Talalayevsky (2000), in order to coordinate a wide range of activities for their clients,
the 3PLs have been using more and more complex information technologies (ITs). In these cases the transmission
of the ‘right information to the right person at the right time so it can be used in real time’ is one of the challenges of

Table 1. Literature on main drivers of logistics activity outsourcinga.

Article Method Locus Drivers identified

Briggs et al. (2010) Survey US Increase flexibility; improve service;

Perçin (2009) Theoretical/case application of a
novel 3PL selection methodology
Huo et al. (2008) Survey
Ross et al. (2007) Theoretical/case application of a
novel 3PL costing methodology
within the offshore industry
increase overall performance
Turkey Increase flexibility; improve service; gain
market knowledge; enhance opera-
tions control; focus on core activities
Hong Kong Increase flexibility; improve service;
reduce costs; increase overall
performance
US Reduce costs; increase flexibility

Note: aA comprehensive literature review encompassing previous years is given in Table A1 (Appendix A).
 International Journal of Production Research 2425

providing logistics services (Youngberg et al. 2009). In particular, Enterprise Resource Planning (ERP) systems
standardise and integrate order-related information, rendering it more reliable for the 3PL’s planning of
transportation and warehousing resources relative to client inventories, and thus making improved operational
performance possible (Chou and Chang 2008). The mastery of sophisticated IT, e.g. by making a variety of
information, including tracking of goods, available over the Internet (Lieb and Lieb 2008), has given 3PLs
prominence in the market. In sum, clients consider technological prowess as a basic item expected of 3PLs
(Lieb 2005).
The recognition of the importance of coordination processes on transportation and warehousing, supply chain
key functions performed by 3PLs, is not new (Ng et al. 1997). The novelty is the appearance of IT applications,
which has transformed the operational mode of these activities and leveraged supply chain performance (Mason
et al. 2003, Ganesh 2005, Stefansson and Lumsden 2009). Transportation and warehousing management systems,
for example, are key technologies used to manage the physical flow of merchandise along the supply chain.
Integrated systems (including transportation management systems, warehousing management systems, and global
inventory visibility via the Internet) may potentially drive costs down and improve client services through the better
matching of resources with demands, thus reducing shipping/receiving lead times, yielding more accurate shipping
and reducing the variability of response times (Mason et al. 2003).
The 3PLs have been adopting certification processes, such as those developed by the International Organization
Downloaded by [Universite Laval] at 18:27 17 September 2015

for Standardization (ISO), more and more valued by shippers. By means of structuring and implementation of
standardised procedures, certification tends to be associated with improved service levels. For example, it has been
empirically shown that ISO 9000 compliance improves the performance of logistical operations, providing positive
results soon after adoption. Better performance translates into shorter lead times for products, and shorter turnover
for cash circulating among suppliers, clients, and service providers (Lo et al. 2009).
According to Zhou et al. (2008), despite the numerous studies on the 3PL sector, only two had attempted to
evaluate the performance of the industry using DEA until that date. This shortcoming clearly suggests a void to be
filled. Min and Joo (2006), for example, applied the technique to a group of six leading US-based 3PLs. The authors
developed a benchmark as a way to identify the 3PLs developing best practices and to allow other 3PLs to emulate
them. According to them, the DEA technique helps to guide financial investments and to assess the impacts of
investments on firm performance. The results indicated that the 3PLs in the US, which ranked among the 25 largest
in 2000, could not be considered as efficient during any period investigated (1999–2002). It was also noted that, in
2001, the fall in the rate of growth in the US manufacturing sector was correlated with a decline in the operational
performance of the 3PLs studied.
Hamdan and Rogers (2007) applied the DEA technique to 3PL warehousing operations. Nineteen warehouses
belonging to a 3PL in the US were studied. The study reflects the importance of warehousing processes for the
sector. For purposes of modelling, inputs that represented work, space, technology, and equipment were chosen, as
well as outputs that represented quantity produced, order fulfillment, and use of space. The analyses were validated
by the 3PL: four of the six warehouses classified as efficient by DEA had already been ranked among the firm’s
highest performers.
In China, Zhou et al. (2008) applied the DEA technique to the 3PL sector. Their main intentions were to
establish a benchmark for the sector and to identify factors that could affect the performance of the 3PLs. After
measurement of the operational performance of the group under study, DEA scores were regressed against four
potential impacting factors by means of Ordinary Least Squares (OLS). Among the main conclusions was the fact
that company size did not necessarily impact 3PL efficiency in a positive way, as would be expected. It was also
discovered that accumulated sales revenues enabled a better use of 3PL resources, and that investments in staff
training were good for personnel retention and for 3PL performance.
More recently, Min and Joo (2009) again assessed the comparative strengths and weaknesses of leading third-
party logistics providers (3PLs) in the USA with respect to their financial efficiencies during the period 2005–2007.
The authors used DEA to measure the financial efficiency of 12 leading 3PLs in the US, relative to their key
competitors, determining that business expansion through mergers and acquisitions can hurt the financial efficiency
of 3PLs due to restructuring and re-branding costs.
These last four papers indicate that the identification of contextual variables (environmental factors) that
significantly affect efficiency is, in general, the most relevant methodological issue to studies that apply DEA to
logistics firms. Only Zhou et al. (2008) managed to regress DEA scores against contextual variables. Other DEA
methodological issues are related to the sample size adequacy in order to avoid a concentration of scores in one
(Tongzon 2001), the proper use of the Tobit regression in order to handle truncated scores in zero and one
 2426 P.F. Wanke

(Turner et al. 2004), and the use of panel data models to adequately regress different efficiency scores, measured
from the same 3PLs over the years, against contextual variables (Greene 2007). Thus far, the scant previous research
that applied DEA to 3PL firms did not observe these issues.
This research differs from previous studies by analysing the Brazilian 3PL sector between 2001 and 2010 using a
two-stage DEA model. In the first stage, DEA is used to calculate efficiency scores for each 3PL firm and, in the
second stage, these scores are used as the dependent variable in the corresponding Tobit regression model for
unbalanced panel data. Contextual variables such as ITs adopted and services provided by the 3PLs constitute the
regressors, further detailed in Section 4.
Before proceeding, however, it is worth noting that Section 3 presents some methodology fundamentals
concerning DEA – returns to scale, input/output orientation, non-parametric frontiers, and scale efficiency – and
the statistical treatment of its scores – by means of Tobit regressions that can be applied to (un)balanced panel
data – in order to support the data analysis and the discussion of results subsequently presented in Section 4.

3. Methodological fundaments
3.1 DEA background
Downloaded by [Universite Laval] at 18:27 17 September 2015

DEA is a non-parametric method first introduced by Charnes et al. (1978). It is based on linear programming and is
used to address the problem of calculating the relative efficiency for a group of Decision Making Units (DMUs)
using multiple measures of inputs and outputs. Given a set of DMUs, inputs and outputs, DEA determines for each
DMU a measure of efficiency obtained as a ratio of weighted outputs to weighted inputs.
Assume s ¼ 1, . . . , S production units, with inputs xTs ¼ ðxs1 , . . . , xsm Þ and outputs yTs ¼ ð ys1 , . . . , ysn Þ. Column
vectors xs and ys form the sth columns of matrices X and Y. Assume further that T ¼ ð1 , . . . , s Þ is a non-negative
vector and eT ¼ ð1, . . . , 1Þ 2 RS is a vector of unit values. The DEA-CCR (Charnes et al. 1978) and the DEA-BCC
(Banker et al. 1984) are shown in Equations (1)–(3) and illustrated in Figure 1:
DEA-CCR DEA-BCC DEA-BCC
input oriented input oriented output oriented
min , min , max ,
, , ,
ð1Þ ð2Þ ð3Þ
s:t: xs  X  0, s:t: xs  X  0, s:t: ys  Y  0,
Y  ys , Y  ys , X  xs ,
  0, e ¼ 1, e ¼ 1:

The DEA-BCC model differs from the CCR model only in the adjunction of the condition e ¼ 1. Together with the
condition   0, this imposes a convexity condition on allowable ways in which the observations for the s DMUs
may be combined (Zhu 2003). As regards the model orientation, if input- or output-oriented, the two measures
provide the same scores under constant returns to scale, but are unequal when varying returns to scale are assumed
(Cooper et al. 2004).

DEA-CCR and BCC models

Output Y
Constant returns-to-scale (CRS/CCR)

M • H is a scale-efficient DMU.
YM
Considering DMU K:
L
YL • XJ / XK is the input saving technical efficiency (VRS);
Varying returns-to-scale (VRS/BCC) • YK / YL is the output increasing technical efficiency (VRS);
H
• X I / X K is the gross scale efficiency (input saving when VRS);
• X I / X J is the pure scale efficiency (input corrected);
• YL / YM is the pure scale efficiency (output corrected).

I J K
YK

XI XJ Xk Source: Odeck and Alkadi (2001)

Input X

Figure 1. Efficiency measurement: DEA-CCR and BCC.

 International Journal of Production Research 2427

Scale inefficiency is, therefore, due to the increase or decrease of returns to scale, which can be computed by
inspecting the sum of the weights under the CCR model specification. If the sum is equal to one, the law of constant
returns to scale prevails; however, if the sum is less than one or more than one, respectively, increasing or decreasing
returns to scale prevail, assuming an input-oriented model. According to Cooper et al. (2007), in order to identify if
the inefficiency of a DMU is due to inefficient operations or to its scale inefficiency, scale efficiency (SE) is
computed using the ratio SE ¼ CCR =BCC . It is important to point out that the maximum value of SE is one,
indicating that the DMU is operating at the most productive scale size.
Compared with the stochastic parametric frontier approach, DEA imposes neither a specific functional
relationship between production outputs and inputs, nor any assumptions on the specific statistical distribution of
the error terms (Cullinane et al. 2006). An efficient frontier is on the boundary of a convex poly tope created in the
space of inputs and outputs, and in which each vertex is an efficient DMU (Dulá and Helgason 1996). Another
feature of DEA is that the relative weights of the inputs and the outputs do not need to be known a priori, that is
these weights are determined as part of the solution of the linear problem (Zhu 2003).
Besides estimating efficiency scores, DEA also provides other relevant information to the inefficient DMUs.
DEA identifies the efficient facet being used for comparison, the combination of the inputs that are being
inefficiently used, and the deviation of specific outputs from the efficient level. It should be noted that efficient
DMUs do not present any slack, this occurs only to inefficient DMUs (Green et al. 1996, Lin and Tseng 2007).
Downloaded by [Universite Laval] at 18:27 17 September 2015

3.2 Two-stage DEA models – treating statistical variations on efficiency scores

The approaches to the statistical treatment of the variations in the scores produced using DEA – CCR, BCC, or SE
– have evolved over the years (see, for example, Banker (1993) and Simar and Wilson (2007)). As a depiction of this
evolution, Cooper et al. (2007) point to the growing number of studies that combine DEA scores, obtained in a first
stage, with multivariate data analysis, such as regression analysis, in a second stage, when these scores are
incorporated in the form of the dependent variable. According to Fried et al. (2002), such two-stage DEA
approaches are an important recognition that environmental factors or contextual variables can significantly
influence efficiency scores. The authors also show that managerial competence (or incompetence) is insufficient to
explain individual variations in efficiency, given that environmental factors, contextual variables, or even statistical
noise can exert some influence over measured performance. The adequate control of these impacts may suggest
possible paths for a DMU to become more efficient (see, for example, Souza et al. (2007)).
Turner et al. (2004) advocate the use of Tobit regression on DEA scores. In general, the basic model for Tobit
regression is similar to that for OLS, however the former assumes a truncated normal distribution in lieu of a
normal distribution and employs maximum likelihood estimation (Greene 2007). Banker and Natarajan (2008)
showed that DEA-based procedures using Tobit regression in the second stage perform as well as the best of the
parametric methods in the estimation of the impact of contextual variables on efficiency. It should be noted,
nevertheless, that the use of other non-parametric tests, such as those presented by Banker and Natarajan (2004)
and Gomes et al. (2009), constitute an alternative used just as commonly as Tobit regressions in similar situations.
Tobit regression can also be applied to (un)balanced panel data (Greene 2007). Generally speaking, panel data
models allow the examination of fixed or random effects of a specific firm or of time periods on efficiency scores
(Park 2005). For random effects models – according to Greene (2007) the model most frequently used – the basic
assumptions are: the random effect ui is the same for all periods and should not be correlated with other regressors;
the angular coefficients are the same for all groups and periods; and "it , the stochastic component of the model, does
not correlate across periods. The functional form of the random effects model is given by
yit ¼ Xit  þ ui þ "it , ð4Þ
where i denotes the group or individual, t denotes the time period, yit denotes the dependent variable, and Xit
denotes the vector of independent variables.

4. Data analysis and discussion of results

This section is structured in three subsections. The first subsection provides the rationale for choosing scale
efficiency as the cornerstone of the data analysis, to the detriment of using CCR or BCC scores individually. Then,
the second subsection presents the inputs, outputs, and contextual variables considered in the two-stage DEA model
 2428 P.F. Wanke

in light of several modelling issues, such as their adequacy in terms of representing a typical 3PL operation in Brazil,
the possibility of reducing the number of inputs/outputs used in order to enhance the discerning power of the
analysis, and their validity as data obtained from secondary sources. The third subsection encompasses the steps
taken throughout the two-stage DEA model. Not only are descriptive statistics on the scores obtained in stage one
and Tobit regression results derived in stage two presented, but also the use of data reduction techniques in
conjunction with DEA is discussed with respect to contextual variables.

4.1 Why focusing on scale efficiency?

As mentioned above, this study intends to determine the principal factors that affect scale efficiency in the Brazilian
3PL industry. Our interest in scale efficiency is not merely to determine whether a particular 3PL is operating at – or
close to – its optimum level, given the set of inputs used and the level of outputs generated, it is also to determine the
objective conditions under which this can take place by analogy with the study by Ross and Droge (2004). In other
words, scale efficiency can be used to determine how close each 3PL of the sample is to its corresponding most
productive scale size and to what extent such a distance is a consequence of coordination processes in the supply
chain: management of information flows, inventory synchronisation mechanisms, and scaling of resources
Downloaded by [Universite Laval] at 18:27 17 September 2015

(Wanke 2003).
In large-scale distribution systems (the typical situation of a 3PL), different coordination processes frequently
lead to different patterns of resource allocation among activities, potentially making the adjustments of the scale to
the operation more flexible (Ross and Droge 2004). In this case, the results of scale efficiency may indicate
opportunities for downsizing (decreasing returns to scale) or consolidation of operations (increasing returns to
scale). For example, depending on alternative uses of information technologies (ITs) and mechanisms to synchronise
and move the inventories by 3PLs, there may be situations in which the warehouse experiences decreasing
(increasing) returns to scale due to its very large (small) size compared with inventory levels, movement of cargoes,
and orders that have been allocated (Ross and Droge 2004). The basic idea is, therefore, to verify the role of these
coordination processes (contextual variables) when computing the scale efficiency of the 3PL, assessing whether, in
fact, the 3PL engenders a more rational allocation of resources (inputs) to the demand (outputs) and, consequently,
an operation close to the most productive scale size, with the availability of real-time information as a cornerstone.

4.2 Inputs, outputs, and contextual variables

The inputs, outputs, and contextual variables used in this study were collected from the special edition dedicated to
the 3PL sector in Revista Tecnologı´stica (2001–2010) published in June or July every year. Their final set is presented
in Table 2. It should be noted that the original variable set was cleaned up, rejecting the variables that were not

Table 2. Inputs, outputs, and categorical variables considered in the analysisa.

Inputs Outputs

Number of employees Number of clients

Total warehouse area Gross revenues
Total owned warehouses
Total client warehouses
Categorical variables

ISO certification Internet queries Stock control Project development Distribution

Packaging ERP Foreign offices Intermodal management Customs clearance
JIT – Just in time Reverse logistics Milk run Kit assembly Door to door
Local operation Regional operation Tracking – Own radio Tracking – Outsourced Tracking – Outsourced
radio satellite
Tracking – Own Routing – Own Inspections support Transfers WMS
satellite

Note: a1 ¼ characteristic present; 0 ¼ characteristic not present (categorical variables only).

 International Journal of Production Research 2429

collected for all the individuals listed in the panel. This was done in order to homogenise the information for the
analysis.
Conducting a secondary analysis of existing data saved the time and resources needed to collect primary data.
However, the benefits of saving time and effort must be weighted against the limitations due to the level of data and
the lack of specificity of the data for the secondary project (Shepard et al. 1999). All the data collected from Revista
Tecnologı´stica are objective measures based on explicit criteria, represented by metric (inputs and outputs) and
nominal scales (most of the contextual variables, with the exception of age). As single-item indicators of objective
measures, the data can be valid and reliable indicators of the variables under consideration (Youngblut and
Casper 1993).
Although the data set provided by Revista Tecnologı´stica might not have been collected in the context of a
theoretical model, a theoretical model can still be identified and applied to the research process and data that are
theoretically consistent can be identified (Zill and Daly 1993, Moriarty et al. 1999). The importance of this step in
secondary analysis cannot be underestimated (Shepard et al. 1999). As with any quantitative method of research,
selection of the variables to be studied must first involve combing through the model to identify critical concepts.
The theoretical concepts are then matched with appropriate variables from the data set.
Therefore, in order to build the DEA models, the four inputs and two outputs available to all 3PLs were used.
Their descriptive statistics are given in Table A2 (see Appendix A). Following the example of previous studies (Zhou
Downloaded by [Universite Laval] at 18:27 17 September 2015

et al. 2008), their measurement units represent resources that are critical not only financially, but also for the
consecution of logistical services.
With respect to inputs, the 3PL’s total number of staff members involved in either strategic or operational
activities is the measure used to represent labour force utilisation. Beyond this, a selection of measures is also
necessary that translate how the 3PLs handle warehousing – the logistics activity that has grown the most up to 2008
in Brazil (Marino 2008). According to the author, in the Brazilian 3PL sector, the availability of warehousing
services is greater than the availability of transportation services. This being the case, the total area of owned
warehouses is used as an input for the model. It is also important to take into account those situations where the
3PL operates the warehouse, although the asset itself belongs to the client (Marino 2008). In the latter case, that
warehouse, which functions as one of the 3PL’s operational resources but not as one of its assets, is computed on the
total of the client’s warehouses, constituting the model’s last input.
Regarding outputs, measures that represent financial and operational aspects are used. Gross revenue is
considered an output, because it portrays the provided services (Min and Joo 2006, 2009). The firm’s total number
of clients, in a similar way, reflects its operational complexity – a large client roster looks good, not only in the
market, but also in terms of suggesting greater ability in the management of different logistics requirements.
Because of the efficient frontier’s geometry, DEA results heavily rely on the set of inputs and outputs used. The
more variables (inputs and outputs) in the DEA, the less discerning the analysis is (Jenkins and Anderson 2003).
This fact demands greater concern for the variable selection process. Given the large number of initial potential
variables to be considered for DEA, several methods have been proposed in the literature, which suggest limiting the
number of variables in relation to the number of DMUs (Wagner and Shimshak 2007). Some studies have suggested
that judgment should be performed by specialists in order to indicate which variables are the most relevant for the
DEA model (Golany and Roll 1989). Other studies have suggested regression analysis, in order to indicate highly
correlated variables as redundant (Lewin et al. 1982), or even the application of DEA to smaller models, in order to
rank the effect of variables on efficiency scores (Wagner and Shimshak 2007).
Therefore, in order to check the possibility of reducing the number of inputs and outputs to be considered in the
analysis, correlation analyses were performed. Table 3 shows the correlation coefficients between the pairs of inputs
and the pair of outputs. They were calculated from the 251 observations – comprising 131 different individuals
distributed over the course of these 10 years – taken in aggregate. Since the serial correlations are relatively low, all
inputs and outputs were kept in the analysis.

4.3 Two-stage DEA

In the first stage, the DEA-CCR and BCC models were executed 10 times using Frontier Analyst 4.0.10, i.e. once for
each year for the period 2001–2010. Table 4 shows the descriptive statistics of the scores computed for the CCR and
BCC models and for the scale efficiency for the years 2001–2010. As expected, the CCR models returned efficiency
scores that were lower than those computed for the BCC models. In other words, the CCR models identified fewer
 2430 P.F. Wanke
Table 3. Correlation coefficient matrix.

Total Total Total

Number warehouse owned client Number Gross
Variable of employees area warehouses warehouses of clients revenue

Number of employees 1.00

Total warehouse area 0.51 1.00
Total owned warehouses 0.34 0.44 1.00
Total client warehouses 0.49 0.39 0.24 1.00
Number of clients 1.00
Gross revenue 0.18 1.00

Table 4. Descriptive statistics of DEA scores by year.

Year

Score 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 All
Downloaded by [Universite Laval] at 18:27 17 September 2015

CCR Average 0.19 0.57 0.43 0.62 0.53 0.45 0.44 0.53 0.28 0.45 0.41
Minimum 0.00 0.15 0.06 0.10 0.20 0.14 0.05 0.09 0.01 0.05 0.00
Maximum 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Standard deviation 0.28 0.33 0.34 0.31 0.30 0.33 0.36 0.33 0.34 0.31 0.35
Coefficient of variation 1.47 0.59 0.80 0.50 0.57 0.74 0.82 0.62 1.23 0.68 0.84
# of efficient DMUs 3 3 4 5 2 1 6 7 7 7 45
% of efficient DMUs 9% 27% 17% 25% 17% 20% 20% 23% 14% 18% 18%
BCC Average 0.70 0.87 0.87 0.80 0.83 0.63 0.65 0.77 0.72 0.78 0.75
Minimum 0.17 0.37 0.47 0.25 0.25 0.31 0.06 0.24 0.11 0.24 0.06
Maximum 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Standard deviation 0.29 0.21 0.19 0.28 0.31 0.25 0.34 0.27 0.34 0.25 0.34
Coefficient of variation 0.42 0.24 0.22 0.35 0.37 0.39 0.52 0.35 0.48 0.33 0.45
# of efficient DMUs 11 10 9 12 11 4 13 11 27 18 129
% of efficient DMUs 33% 91% 39% 60% 92% 80% 43% 37% 55% 47% 51%
SE Average 0.24 0.65 0.48 0.78 0.65 0.71 0.63 0.70 0.39 0.59 0.54
Minimum 0.01 0.21 0.06 0.27 0.24 0.26 0.17 0.09 0.02 0.10 0.01
Maximum 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Standard deviation 0.28 0.32 0.34 0.26 0.26 0.32 0.30 0.32 0.36 0.33 0.35
Coefficient of variation 1.13 0.49 0.70 0.33 0.39 0.46 0.48 0.46 0.92 0.56 0.65
# of efficient DMUs 3 3 4 6 2 1 6 7 7 14 53
% of efficient DMUs 9% 27% 17% 30% 17% 20% 20% 23% 14% 37% 21%
Total DMUs 33 11 23 20 12 5 30 30 49 38 251
# of DMUs – CRS 11 10 9 12 11 4 13 11 7 9 93
# of DMUs VRS – Increasing – 0 1 1 0 0 1 1 41 28 73
# of DMUs VRS – Decreasing 22 1 13 7 1 1 16 18 1 1 79

Note: CRS, constant returns to scale; VRS, variable returns to scale.

efficient 3PLs than the BCC models for each year. This result is unsurprising, given that the CCR model assumes a
production technology with constant (linear) returns of scale. The BCC model, on the other hand, assumes variable
returns to scale, which more closely parallels reality since they reflect the technical efficiency of different DMUs (see
Figure 1). In addition, it can be seen that very few 3PLs operate at the most productive scale size.
In the second stage, traditional characteristics, ITs, and services commonly offered by 3PLs in Brazil (see
Table 2) were considered as regressors, in order to identify the determinants of scale efficiency of 3PLs operating
nationally. Such characteristics, ITs, and services are the control variables of this study, since they comprise neither
process inputs nor products, but rather their attributes, in a total of 25 contextual variables. These variables are
terminal, i.e. they assume the value of one if the observation has the mentioned characteristic and zero otherwise. It
is understood that k  1 dummy variables are required to represent a variable with k categories (Levine et al. 2007).
 International Journal of Production Research 2431
Table 5. Rotated component matrixa.

FACTOR 1 – Stock and warehousing-related ITs and services Stock control 0.74
Packaging 0.74
Kit assembly 0.80
Inspections support 0.61
WMS 0.58
FACTOR 2 – Owned tracking and routing ITs Routing – Own 0.85
Tracking – Own satellite 0.88
Tracking – Own radio 0.79
FACTOR 3 – Classical transportation-related services Distribution 0.78
Door to door 0.77
Transfers 0.81
Reverse logistics 0.50
FACTOR 4 – Express logistics-related ITs and services Tracking – Outsourced satellite 0.54
Tracking – Outsourced radio 0.65
Just in time 0.59
Milk run 0.50
Intermodal management 0.57
ERP 0.50
FACTOR 5 – Foreign operations and ISO certification ISO certification 0.50
Downloaded by [Universite Laval] at 18:27 17 September 2015

Foreign offices 0.76

Customs clearance 0.59
FACTOR 6 – Age Age 0.59

Note: aKMO ¼ 0.696. Chi-square ¼ 1730.282 (sig. ¼ 0.000). All factor loads greater than 0.50 should be interpreted.

The base category is the absence of a given characteristic, IT, or service. Besides these variables, the only exception
should be mentioned: age of the 3PL, measured in months (metric scale).
Given the large number of potential contextual variables to be considered in the second stage, data reduction
techniques assume particular relevance here. Several authors have used some of these techniques together with
DEA. Adler and Golany (2001) and Adler and Berechman (2001), for instance, employed principal component
analysis. The use of factor analysis was proposed by Vargas and Bricker (2000) and implemented by Jenkins and
Anderson (2003) and Nadimi and Jolai (2008). Specifically, factor analysis is an appropriate procedure for data
reduction based on observed variables and on existing theoretical constructs (Hair et al. 2005).
Therefore, in order to make the concept of coordination processes and information technologies operational –
building it upon these 25 contextual variables related to 3PL’s characteristics, ITs, and services – a factor analysis
with Varimax standardised rotation was conducted with the use of SPSS 15.0. Table 5 presents the six factors related
to coordination processes and information technologies.
Finally, LIMDEP 9.0 econometric software was used to carry out the Tobit regression on the unbalanced
panel data, using the random effects model. The results were adjusted due to the heteroscedasticity produced by
groups of different size (Greene 2007). The range 0.05–0.10 was established with respect to the acceptable level
of significance, as has been customary in exploratory research studies on logistics (Mentzer and Flint 1997,
Wanke and Hijjar 2009). Table 6 shows the Tobit regression results for each one of these six factors. The
results presented in Table 6 confirm the impact of coordination processes on the supply chain and, in particular,
the impact of ITs on increased scale efficiency for Brazilian 3PLs. The adoption of express logistics-related ITs
and services (FACTOR 4) merit attention: radio and satellite tracking (outsourced), ERP, just in time, milk run,
and intermodal management.
Embedded within these results, it should be noted that inventory-related coordination processes, such as just in
time and milk run, presented significant positive impacts on efficiency. A possible justification for this effect is the
fact that inventory-related coordination processes allow for a greater integration of client product flow with the 3PL
transportation and warehousing resources needed for their movement.
It should also be noted that the age (FACTOR 6) of the 3PL also has a positive effect on scale efficiency.
In addition to the experience accumulated from operating for a longer time in the market, we should also take
into account the fact that the relationship between the contracting company and the 3PL tends to become more
focused, thereby allowing for a better tailoring of resources to client exigencies, as advocated by Bhatnagar
et al. (1999).
 2432 P.F. Wanke
Table 6. Tobit regression results (random effects)a.

Variable Coefficient Standard error b/St. err. P[Z 4 z] Mean of X

FACTOR 1 0.023 0.041 0.558 0.576 0.000

FACTOR 2 0.052 0.043 1.192 0.283 0.000
FACTOR 3 0.016 0.290 0.546 0.585 0.000
FACTOR 4 0.112 0.032 3.491 0.000 0.000
FACTOR 5 0.057 0.039 1.442 0.149 0.000
FACTOR 6 0.071 0.028 2.555 0.011 0.000
Sigma (v) 0.342 0.020 16.910 0.0000
Sigma (u) 0.904 0.046 16.546 0.0000

Note: aMcFadden’s pseudo-R2 ¼ 0.220. Chi-squared ¼ 125.182. Degrees of freedom ¼ 1. Prob[chi-

squared 4 value] ¼ 0.0000000.
Unbalanced panel contains 131 individuals.

5. Managerial implications
As competition in the 3PL industry has intensified over the last decade, today’s 3PLs are faced with daunting
challenges of continuously improving their efficiency levels and competitiveness. In this sense, this study lends a
Downloaded by [Universite Laval] at 18:27 17 September 2015

contribution of a practical nature to the 3PL sector in Brazil, supporting not only 3PLs but also shippers in their
decision-making processes. More precisely, 3PL managers may use its general results as guidance for future steps
towards higher levels of scale efficiency. ‘What ITs should be developed (acquired) first?’ and ‘what kinds of logistics
services should be offered to shippers?’ constitute examples of questions that may direct 3PLs through a shorter path
to the most productive scale size, helping them to establish a business plan or an action course over time. The
prominent role of ‘express logistics-related ITs and services’ is evident when compared with other factors.
Third-party logistics managers could also use the scale developed for the measurement of scale efficiency as a
self-assessment tool for benchmarking against industry practice. ‘What is my current scale efficiency level?’ and ‘is
my company positioned above or below average scale efficiency?’ are questions that could be answered by using the
results presented in Tables 5 and 6. Besides the self-assessment issue, when the impact of adopting additional ITs
and/or providing additional services on scale efficiency can be marginally evaluated, business expansion options
through mergers and acquisitions can be also assessed by means of the resultant service/IT mix obtained from the
conjoint operation of both companies.
The findings of this study may also serve as a valuable tool for shippers to benchmark their logistics services
providers against each other. Even though no link among scale efficiency, costs, and service levels is claimed in the
evidence presented and discussed in this paper, it serves as an indication of the directions shippers should take when
hiring 3PL services. The basic underlying idea is that 3PLs with higher levels of scale efficiency may simultaneously
achieve lower costs and higher service levels, thus benefiting shippers in terms of competitive advantage.

6. Conclusions
In this study, a two-stage DEA model was used both to evaluate the scale efficiency of 3PLs in Brazil and to identify
their main determinants from 2001 to 2010. The results corroborate evidence in the literature that coordination
mechanisms in the supply chain, supported by the availability of real-time information and inventory
synchronisation, favour a more rational allocation of resources (inputs) to client demands (outputs). In other
words, results indicate that these coordination mechanisms may favour an operation that is close to the most
productive scale size.
The contribution of this paper is two-fold. On the theoretical side, a valuable scale for the measurement of scale
efficiency has been built and validated, representing an index towards the most productive scale size. On the other
hand, the managerial implication of this possibility of measuring the scale efficiency of 3PLs is that it may be used as
a basis for segmenting the market met by shippers. For instance, does a given 3PL have higher scale efficiency than
another one? This may imply different concerns in terms of service offer, negotiation conditions, trade-offs between
costs and service levels, etc.
Since a range of variables embedded within factors were statistically validated, areas where there is space for the
development of future studies are revealed. For example, future studies may search to discriminate several aspects of
express logistics services and information technologies based on different characteristics of 3PLs. That is, a clustering
 International Journal of Production Research 2433

approach might be used to identify segments of 3PLs according to characteristics in terms of services and information
technologies provided. Future studies may also focus on the 3PL efficiency frontier by incorporating other
approaches, such as Stochastic Frontier Analysis (SFA), in order to cross-check with DEA results.
Finally, the fact of working with secondary data instead of primary data brings certain limitations to this work,
mainly with respect to the set of inputs and outputs used in the analysis, which may not cover all aspects relevant to
building an efficiency frontier.

References

Adler, N. and Berechman, J., 2001. Measuring airport quality from the airlines’ viewpoint: an application of data envelopment
analysis. Transport Policy, 8, 171–181.
Adler, N. and Golany, B., 2001. Evaluation of deregulated airline networks using data envelopment analysis combined with
principal component analysis with an application to Western Europe. European Journal of Operations Research, 132,
260–273.
Arroyo, P., Gaytan, J., and de Boer, L., 2006. A survey of third party logistics in Mexico and a comparison with reports on
Europe and USA. International Journal of Operations & Production Management, 26 (6), 639–667.
Banker, R.D., 1993. Maximum likelihood, consistency and DEA: a statistical foundation. Management Science, 39 (10),
Downloaded by [Universite Laval] at 18:27 17 September 2015

1265–1273.
Banker, R.D., Charnes, A., and Cooper, W.W., 1984. Some models for the estimation of technical and scale inefficiencies in data
envelopment analysis. Management Science, 30 (9), 1078–1092.
Banker, R.D. and Natarajan, R., 2004. Statistical tests based on DEA efficiency scores. In: W.W. Cooper, L.M. Seiford and
J. Zhu, eds. Handbook on data envelopment analysis. Boston: Kluwer International Series, 299–321.
Banker, R.D. and Natarajan, R., 2008. Evaluating contextual variables affecting productivity using data envelopment analysis.
Operations Research, 56, 48–58.
Bardi, E.J. and Tracey, R., 1991. Transportation outsourcing: a survey of US practices. International Journal of Physical
Distribution and Logistics Management, 21 (3), 15–21.
Bathnagar, R., Sohal, A.S., and Millen, R., 1999. Third party logistics services: a Singapore perspective. International Journal of
Physical Distribution & Logistics Management, 29 (9), 569–587.
Boyson, S., et al., 1999. Managing effective third party logistics relationships: what does it take? Journal of Business Logistics,
20 (1), 73–100.
Briggs, E., Landry, T., and Daugherty, P., 2010. Investigating the influence of velocity performance on satisfaction with third
party logistics service. Industrial Marketing Management, 39 (4), 640–649.
Buck, D., 1988. Changing to contract distribution. Management Decision, 26 (5), 42–47.
Carbone, V. and Stone, M., 2005. Growth and relational strategies used by the European logistics service providers: rationale
and outcomes. Transportation Research. Part E, Logistics & Transportation Review, 41E (6), 495–510.
Cavinato, J.L., 1989. The logistics of contract manufacturing. International Journal of Physical Distribution and Logistics
Management, 19 (1), 13–20.
CEL, 2009. Panorama logı´stico CEL/COPPEAD – Terceirização logı´stica no Brasil. Rio de Janeiro: COPPEAD/UFRJ.
Charnes, A., Cooper, W.W., and Rhodes, E., 1978. Measuring efficiency of decision-making units. European Journal of
Operational Research, 2, 429–444.
Chou, S. and Chang, Y., 2008. The implementation factors that influence the ERP (enterprise resource planning) benefits.
Decision Support Systems, 46, 149–157.
Cook, W.D. and Seiford, L.M., 2009. Data envelopment analysis (DEA) – thirty years on. European Journal of Operational
Research, 192, 1–17.
Cooper, W.W., Seiford, L.M., and Tone, K., 2007. Data envelopment analysis: a comprehensive text with models, applications,
references and DEA-solver software. New York: Springer.
Cooper, W.W., Seiford, L.M., and Zhu, J., 2004. Handbook on data envelopment analysis. Boston: Kluwer Academic Publishers.
Cullinane, K., et al., 2006. The technical efficiency of container ports: comparing data envelopment analysis and stochastic
frontier analysis. Transportation Research Part A, 40, 354–374.
Dulá, J.H. and Helgason, R.V., 1996. A new procedure for identifying the frame of the convex hull of a finite collection of points
in multidimensional space. European Journal of Operational Research, 92, 352–367.
Fernie, J., 1989. Contract distribution in multiple retailing. International Journal of Physical Distribution and Logistics
Management, 19 (7), 2–35.
Figueiredo, K.F. and Mora, D.M., 2009. A segmentação dos operadores logı́sticos no mercado brasileiro de acordo com suas
capacitações para oferecer serviços. RAC-Eletrônica, 3 (1), 123–141.
Fleury, P.F., 2000. Vantagens competitivas e estratégicas do uso de operadores logı́sticos. In: P.F. Fleury, K.F. Figueiredo and
P.F. Wanke, eds. Logı´stica empresarial: a perspectiva Brasileira. São Paulo: Atlas, 133–142.
 2434 P.F. Wanke

Fleury, P.F. and Ribeiro, A.F.M., 2003. A indústria de provedores de serviços logı́sticos no Brasil. In: P.F. Fleury,
K.F. Figueiredo and P.F. Wanke, eds. Logı´stica e gerenciamento da cadeia de suprimentos: planejamento do fluxo e dos
recursos. São Paulo: Atlas, 302–312.
Fried, H.O., et al., 2002. Accounting for environmental effects and statistical noise in Data Envelopment Analysis. Journal of
Productivity Analysis, 17, 157–174.
Ganesh, V., 2005. A framework for evaluating third-party logistics. Association for Computing Machinery. Communications of the
ACM, 48 (1).
Golany, B. and Roll, Y., 1989. An application procedure for DEA. Omega, 17, 237–250.
Gomes, E.G., et al., 2009. Efficiency and sustainability assessment for a group of farmers in the Brazilian Amazon. Annals of
Operations Research, 69 (1), 167–182.
Green, R., Doyle, J., and Cook, W.D., 1996. Efficiency bounds in data envelopment analysis. European Journal of Operational
Research, 89 (3), 482–490.
Greene, W.H., 2007. LIMDEP version 9.0 – Econometric modeling guide. New York: Econometric Software, Inc.
Hair, J.F., et al., 2005. Análise multivariada de dados. Porto Alegre: Bookman.
Hamdan, A. and Rogers, K.J., 2007. Evaluating the efficiency of 3PL logistics operations. International Journal of Production
Economics, 113 (1), 235–244.
Huo, B., et al., 2008. Understanding drivers of performance in the 3PL industry in Hong Kong. International Journal of
Operations & Production Management, 28 (8), 772–800.
Jenkins, L. and Anderson, L., 2003. A multivariate statistical approach to reducing the number of variables in data envelopment
Downloaded by [Universite Laval] at 18:27 17 September 2015

analysis. European Journal of Operational Research, 147, 51–61.

Knemayer, A.M. and Murphy, P.R., 2004. Evaluating the performance of third-party logistics arrangements: a relationship
marketing perspective. Journal of Supply Chain Management, 40 (1), 35–51.
La Londe, B.J. and Maltz, A.B., 1992. Some propositions about outsourcing the logistics functions. The International Journal of
Logistics Management, 3 (1), 1–11.
Levine, D.M., et al., 2007. Estatı´stica: teoria e aplicações. São Paulo: LTC.
Lewin, A., Morey, R., and Cook, T., 1982. Evaluating the administrative efficiency of courts. Omega, 10, 401–411.
Lewis, I. and Talalayevsky, A., 2000. Third-party logistics: leveraging information technology. Journal of Business Logistics,
21 (2), 173–185.
Lieb, R.C., 2005. The 3PL industry: where it’s been, where it’s going. Supply Chain Management Review, 9 (6), 20–27.
Lieb, R.C., 2008. The North American third-party logistics industry in 2007: the provider CEO perspective. Transportation
Journal, 47 (2), 39–53.
Lieb, R.C. and Bentz, B.A., 2005. The use of third-party logistics services by large American manufacturers: the 2004 survey.
Transportation Journal, 44 (2), 5–15.
Lieb, R.C. and Lieb, K.J., 2008. Why 3PLs need to build their brand? Supply Chain Management Review, 12 (8), 46–52.
Lieb, R.C. and Randall, H.L., 1996. A comparison of the user of third-party logistics services by large American manufacturers,
1991, 1994 and 1995. Journal of Business Logistics, 17 (1), 55–62.
Lin, L.C. and Tseng, C.C., 2007. Operational performance evaluation of major container ports in the Asia-Pacific region.
Maritime Policy & Management, 34 (6), 535–551.
Lo, C.K.Y., Yeung, A.C.L., and Cheng, T.C.E., 2009. ISO 9000 and supply chain efficiency: empirical evidence on inventory and
account receivable days. International Journal of Production Economics, 118, 367–374.
Marino, S., 2008. Mercado promissor. Revista Tecnologı´stica, 19 (151), 68–73.
Mason, S.J., et al., 2003. Integrating the warehousing and transportation functions of the supply chain. Transportation Research
Part E – Logistics & Transportation Review, 39E (2), 141–159.
Mentzer, J.T. and Flint, D.J., 1997. Validity in logistics research. Journal of Business Logistics, 18 (1), 199–216.
Min, H. and Joo, S.J., 2006. Benchmarking the operational efficiency of third party logistics providers using data envelopment
analysis. Supply Chain Management: An International Journal, 11 (3), 259–265.
Min, H. and Joo, S.J., 2009. Benchmarking third-party logistics providers using data envelopment analysis: an update.
Benchmarking: An International Journal, 16 (5), 572–587.
Moriarty, H.J., et al., 1999. Issues to consider when choosing and using large national databases for research of families. Western
Journal of Nursing Research, 21 (2), 143–153.
Nadimi, R. and Jolai, F., 2008. Joint use of Factor Analysis (FA) and Data Envelopment Analysis (DEA) for
ranking of data envelopment analysis. International Journal of Mathematical, Physical and Engineering Sciences,
2 (4), 218–222.
Ng, B., Ferrin, B.G., and Pearson, J.N., 1997. The role of purchasing/transportation in cycle time reduction. International
Journal of Operations & Production Management, 17 (6), 574–591.
Odeck, J. and Alkadi, A., 2001. Evaluating efficiency in the Norwegian bus industry using Data Envelopment Analysis.
Transportation, 28, 211–232.
Panayides, P.M., 2007. Effects of organizational learning in third-party logistics. Journal of Business Logistics, 28 (2), 133–157.
 International Journal of Production Research 2435

Panayides, P.M., et al., 2009. A critical analysis of DEA applications to seaport economic efficiency measurement. Transport
Reviews, 29, 183–206.
Park, H.M., 2005. Linear regression models for panel data using SAS, Stata, LIMDEP, and SPSS [online]. Available from:
www.indiana.edu/statmath [Accessed 15 April 2005].
Perçin, S., 2009. Evaluation of third-party logistics (3PL) providers by using a two-phase AHP and TOPSIS methodology.
Benchmarking: an International Journal, 16 (5), 588–604.
Rafiq, M. and Jaafar, H., 2007. Measuring customers’ perceptions of logistics service quality of 3PL service providers. Journal of
Business Logistics, 28 (2), 159–175.
Rao, K. and Young, R.R., 1994. Global supply chains: factors influencing outsourcing of logistics functions. International
Journal of Physical Distribution and Logistics Management, 24 (6), 11–19.
Razzaque, A.M. and Sheng, C.C., 1998. Outsourcing of logistics functions: a literature survey. International Journal of Physical
Distribution and Logistics Management, 28 (2), 89–107.
Ross, A.D. and Droge, C., 2004. An analysis of operations efficiency in large-scale distribution systems. Journal of Operations
Management, 21, 673–688.
Ross, A.D., Vaidyanathan, J., and Powell, R., 2007. Optimizing 3PL service delivery using a cost-to-serve and action research
framework. International Journal of Production Research, 45 (1), 83–101.
Sahay, B.S. and Mohan, R., 2006. 3PL practices: an Indian perspective. International Journal of Physical Distribution and
Logistics Management, 36 (9), 666–689.
Sheffi, Y., 1990. Third party logistics: present and future prospects. Journal of Business Logistics, 11 (2), 27–39.
Downloaded by [Universite Laval] at 18:27 17 September 2015

Shepard, M.P., et al., 1999. Conceptual and pragmatic considerations in conducting a secondary analysis – an example from
research of families. Western Journal of Nursing Research, 21 (2), 154–167.
Simar, L. and Wilson, P.W., 2007. Estimation and inference in two-stage, semiparametric models of production processes.
Journal of Econometrics, 136 (1), 31–64.
Sink, H.L. and Langley Jr, J.C., 1997. A managerial framework for the acquisition of third-party logistics services. Journal of
Business Logistics, 18 (2), 163–189.
Sink, H.L., Langley Jr, C.J., and Gibson, B.J., 1996. Buyer observations of US third party logistics market. International Journal
of Physical Distribution and Logistics Management, 26 (3), 38–46.
Sohail, M.S., Bhatnagar, R., and Sohal, A.S., 2006. A comparative study on the use of third party logistics services by Singaporean and
Malaysian firms. International Journal of Physical Distribution and Logistics Management, 36 (9), 690–701.
Souza, G.S. and Staub, R.B., 2007. Two-stage inference using data envelopment analysis efficiency measurements in univariate
production models. International Transactions in Operational Research, 14, 245–258.
Souza, G.S., et al., 2007. Economic efficiency of Embrapa’s research centers and the influence of contextual variables. Pesquisa
Operacional, 27, 15–26.
Stefansson, G. and Lumsden, K., 2009. Performance issues of smart transportation management systems. International Journal
of Productivity and Performance Management, 58 (1), 55–70.
Szymankiewicz, J., 1994. Contracting out or selling out? Survey into the current issues concerning the outsourcing of
distribution. Logistics Management, 7 (1), 28–35.
Tongzon, J., 2001. Efficiency measurement of selected Australian and other international ports using data envelopment analysis.
Transportation Research Part A, 35, 107–122.
Turner, H., Windle, R., and Dressner, M., 2004. North American containerport productivity: 1984–1997. Transportation
Research Part E, 40, 339–356.
Van Laarhoven, P., Berglund, M., and Peters, M., 2000. Third-party logistics in Europe – five years later. International Journal of
Physical Distribution and Logistics Management, 30 (5), 425–442.
Vargas, C. and Bricker, D., 2000. Combining DEA and factor analysis to improve evaluation of academic departments given
uncertainty about the output constructs. Working Paper, Department of Industrial Engineering, University of Iowa.
Wagner, J.M. and Shimshak, D.G., 2007. Stepwise selection of variables in data envelopment analysis: procedures and
managerial perspectives. European Journal of Operational Research, 180, 57–67.
Wanke, P.F., 2003. Logı́stica, gerenciamento de cadeias de suprimento e estratégia logı́stica. In: P.F. Fleury, K.F. Figueiredo and
P.F. Wanke, eds. Logı´stica e gerenciamento da cadeia de suprimentos: planejamento do fluxo e dos recursos. São Paulo:
Atlas, 27–47.
Wanke, P.F., Arkader, R., and Hijjar, M.F., 2007. Logistics sophistication, manufacturing segments and the choice of logistics
providers. International Journal of Operations & Production Management, 27 (5), 542–559.
Wanke, P.F. and Hijjar, M.F., 2009. Exportadores brasileiros: estudo exploratório das percepções sobre a qualidade da
infraestrutura logı́stica. Produção, 19 (1), 143–162.
Wilding, R. and Juriado, R., 2004. Customer perceptions on logistics outsourcing in the European consumer goods industry.
International Journal of Physical Distribution & Logistics Management, 34 (8), 28–48.
Yang, Z., 2006. A two-stage DEA model to evaluate the overall performance of Canadian life and health insurance companies.
Mathematical and Computer Modelling, 43 (7/8), 910–919.
 2436 P.F. Wanke

Yellepeddi, S. and Rajagopalan, S., 2005. An interface modeling framework for third party logistics services assessment. IIE
Annual Conference Proceedings, 1–10.
Youngberg, E., Olsen, D., and Hauser, K., 2009. Determinants of professionally autonomous end user acceptance in an
enterprise resource planning system environment. International Journal of Information Management, 29 (2), 138–144.
Youngblut, J.M. and Casper, G.R., 1993. Focus on psychometrics: single-item indicators in nursing research. Research on
Nursing in Health, 16, 459–465.
Zhou, G., et al., 2008. Evaluating the comparative efficiency of Chinese third-party logistics providers using data envelopment
analysis. International Journal of Physical Distribution & Logistics Management, 38 (4), 262–279.
Zhu, J., 2003. Quantitative models for performance evaluation and benchmarking: data envelopment analysis with spreadsheets and
DEA excel solver. New York: Springer.
Zill, N. and Daly, M., 1993. Researching the family: a guide to survey and statistical data in US families. Washington, DC: US
Department of Health and Human Services.

Appendix A

Table A1. Outsourcing drivers.

Downloaded by [Universite Laval] at 18:27 17 September 2015

Article Method Locus Drivers identified

Sahay and Mohan (2006)
Sohail et al. (2006)
Wilding and Juriado (2004)
Van Laarhoven et al. (2000)
Bhatnagar et al. (1999)
Boyson et al. (1999)
Razzaque and Sheng (1998)
Sink and Langley (1997)
Sink et al. (1996)
Lieb and Randall (1996)
Szymankiewicz (1994)
Rao and Young (1994)
Survey with 2002 top India Cost reduction; core competence; service
500 organisations in competence; improved return of assets;
India improved inventory turns; improved
productivity
Survey, multi-industry Malaysia Time saving; cost saving; improved service
Survey with consumer Europe Tap into higher competences; flexibility;
goods companies cost savings; focus on core business;
avoiding investment; expansion to new
markets
Interview-based survey Europe Cost reduction; service improvement; stra-
with large shippers tegic flexibility; focus on core activities;
change implementation
Survey, multi-industry, Singapore Cost savings; flexibility; focus on core
list of registered com- business; customer satisfaction (service);
panies in Singapore employee morale; productivity improve-
ment; access to up-to-date technology
Survey with transporta- US Cost savings or revenue-enhancing poten-
tion and distribution tial; focus on core business; solving a
subscribers problem area; supply chain redesign
Theoretical Expansion to foreign markets; use of JIT
practices and operations complexity; new
systems and technology; flexibility;
retrenchment to core business; M&A
Survey, multi-industry, US and Europe Cost reduction; increase flexibility;
large shippers improved service; reduce headcount;
focus on core competence; reduce capital
expenditures; availability of expertise; IT
Focus group with multi- US Corporate costs/headcount reduction;
industry shipper product/market line expansion, espe-
representatives cially internationally; increasing
customer demands; need to preserve
capital or reduce risk
Survey with Fortune 500 US Cost reduction; access to data; improve
shippers operations; improve customer services;
focus on core competence; flexibility
Survey with large UK Increase flexibility; improved service;
shippers reduce cost; avoid investment
Case studies by means of US and Europe Focus on core activities; service; cost
personal interviews savings; downsizing; risk and control;
with 15 Fortune 500 systems and IT; market intelligence
shippers

(continued )
 International Journal of Production Research 2437
Table A1. Continued.

Article Method Locus Drivers identified

La Londe and Maltz (1992) Survey with members of US Space requirements; change in logistics
the CLM and WERC management; labour costs; new markets
– multi-industry and products; instituting a quality pro-
gram; M&A
Bardi and Tracey (1991) Survey with members of US Cost reduction; need for specialised exper-
the CLM – multi- tise; asset reduction; expanding workload
industry
Sheffi (1990) Theoretical Focus on core business; better transporta-
tion solution (service); cost savings;
technical expertise and IT; better
equipped logistics services
Cavinato (1989) Theoretical US Cost reduction
Fernie (1989) Interview survey w/dis- US Service; use of assets; need for specialised
tribution directors expertise
from major multiple
retail groups
Buck (1988) Theoretical UK Flexibility; concentrate on core business
Downloaded by [Universite Laval] at 18:27 17 September 2015

Table A2. Inputs and outputs.

Inputs Outputs

Total Total Total

Descriptive Number warehouse owned client Number Gross
Year statistics of employees area warehouses warehouses of clients revenue

2001 Average 615 196,889 6 4 143 191,375,758

Minimum 5 3400 1 1 1 800,000
Maximum 4379 786,454 28 25 3200 5,000,000,000
Standard deviation 995 203,215 5 5 566 865,159,325
Coefficient of variation 1.6 1.0 0.9 1.2 4.0 4.5
2002 Average 621 261,992 6 3 292 50,281,818
Minimum 25 22,000 2 1 3 4,700,000
Maximum 3600 1,132,117 17 10 2895 203,000,000
Standard deviation 1067 316,889 4 3 864 71,965,948
Coefficient of variation 1.7 1.2 0.7 0.8 3.0 1.4
2003 Average 348 125,550 4 3 64 32,786,957
Minimum 25 2100 1 1 3 1,200,000
Maximum 4000 441,000 22 13 650 237,000,000
Standard deviation 817 123,163 4 3 146 57,819,706
Coefficient of variation 2.3 1.0 1.1 0.9 2.3 1.8
2004 Average 928 177,589 6 5 37 77,065,500
Minimum 25 3600 1 1 3 1,700,000
Maximum 4765 550,000 20 18 288 444,030,000
Standard deviation 1355 155,497 5 5 65 115,680,282
Coefficient of variation 1.5 0.9 0.8 1.0 1.8 1.5
2005 Average 1598 246,765 6 7 78 143,866,667
Minimum 115 3600 1 1 3 7,000,000
Maximum 5692 1,224,015 21 26 310 430,000,000
Standard deviation 1791 332,940 6 7 106 152,910,576
Coefficient of variation 1.1 1.3 0.9 1.0 1.4 1.1

(continued )
 2438 P.F. Wanke
Table A2. Continued.

Inputs Outputs

Total Total Total

Descriptive Number warehouse owned client Number Gross
Year statistics of employees area warehouses warehouses of clients revenue

2006 Average 1521 283,600 8 12 27 112,630,400

Minimum 90 42,000 3 2 10 152,000
Maximum 2786 600,000 20 25 68 299,000,000
Standard deviation 1117 256,679 7 10 24 123,090,402
Coefficient of variation 0.7 0.9 0.9 0.8 0.9 1.1
2007 Average 991 232,780 7 7 51 83,900,000
Minimum 26 3600 1 1 3 5,000,000
Maximum 8651 842,200 33 37 387 443,100,000
Standard deviation 1777 222,251 7 8 93 114,583,582
Coefficient of variation 1.8 1.0 1.0 1.2 1.8 1.4
2008 Average 623 174,788 9 6 70 80,736,667
Downloaded by [Universite Laval] at 18:27 17 September 2015

Minimum 29 18,500 1 1 3 3,000,000

Maximum 3450 733,000 42 35 670 386,200,000
Standard deviation 779 184,130 9 8 133 94,917,846
Coefficient of variation 1.2 1.1 1.0 1.3 1.9 1.2
2009 Average 1123 206,025 8 6 357 153,147,755
Minimum 35 6000 1 1 3 1,800,000
Maximum 11,841 1,170,675 36 38 12,000 1,300,000,000
Standard deviation 2093 267,438 9 8 1743 255,503,063
Coefficient of variation 1.9 1.3 1.1 1.5 4.9 1.7
2010 Average 1469 236,014 11 5 634 185,016,316
Minimum 22 8500 1 1 1 1,400,000
Maximum 12,018 1,170,675 47 35 12,000 1,600,000,000
Standard deviation 2554 257,746 13 8 2189 334,294,680
Coefficient of variation 1.7 1.1 1.2 1.4 3.5 1.8