Logistics systems performance

indicator
Lecture delivered by
S P Sarmah
 Logistics Performance

• While carrying out Logistics activities costs are

incurred.
• The cost of logistics activities are the Financial
resources consumed by the company;
• It is divided into fixed and variable costs.
• Main cost categories of Logistics systems is shown in
the next slide
Main cost categories Fixed Costs Variable costs
Storage Costs Administrative Costs Insurance policies
Running cost of Deterioration Costs
storage systems Obsolescence costs
Opportunity costs
Operational
Administrative costs Loading and unloading
Management costs
of issuing and goods costs
computing orders Movement cost
Packaging costs
Stock control and
management costs
Transport costs Depreciation of
transport Insurance cost
Rental means of Variable transport costs
transport
Lost sale
Stock out costs -----
Loss of customer
Loss of image
Plant and equipment
Rates of plant Rental fees (variable
cost
devaluation according to volume)
 Service Level

• Service level encompasses the overall degree of customer

satisfaction;
• It depends on numerous factors (marketing mix), connected to
the product characteristics:
> price;
> promotional offers;
> mode of distribution.
• Possibility to quantify the service level by using suitable
indicators.
• Company profits directly connected to the service level offered
to customers.
 Service Level(contd.)

(1) Minimize costs in a reference time horizon (e.g. a

year), while keeping the service level unchanged.

(2) Determine the optimal service level for maximum

profit (difference between revenues and costs) in a
reference period;
Analysis of a logistics activity and Measures of
the service level
• Order-cycle time: Time interval from the
issuing of an order (or request of service) to
the delivery of the product (or completion of
the service);
Analysis of a logistics activity and Measures of
the service level
components of the order-cycle time:
> order processing time (checking for errors in
the order, preparation of the shipping
documents, updating of the store inventory etc.);
> availability check of the products in the
warehouse;
> assembly time of the products making up the
order (withdrawal from a storage point or
creation of packaging for transport);
> shipping time (movement of products from the
storage point to delivery point, including loading
and unloading of the goods).
 Measures of the service level

• Coefficient of variation
➢ Ratio of the standard deviation and mean of the order-
cycle time;
➢ It is the measure of the reliability of the service
offered to the customers
➢ The lower the value, the greater the reliability.
 Measures of the service level (contd.)

Other possible measures

➢ Percentage of dispatches completed within a
prescribed time (level of transport efficiency)
➢ Percentage of orders dispatched within a prescribed
time (integrity, precision and completeness of the
orders).
Example:
• The company has an order cycle time for its product that
depends on two elements: the assembly time and transport
time.

• Five hundred observations on assembly time and 252 recorded

transport time are available.

• The minimum observed value of assembly time is 2 days and

maximum is 15 days.

• Again minimum transport time is 6 days and maximum is 13

days.
 Historical data at Assembly Contd..

No. of Observations (ℎ𝑖 ) Time (𝑥𝑖 )

1 2
4 3
4 4
18 5
38 6
56 7
64 8
96 9
72 10
68 11
41 12
18 13
12 14
2 15
1 16
Sample days: 15
 Contd..
Historical data at Transport

No. of Observations (ℎ𝑖 ) Time (𝑦𝑖 𝑖 )

19 7
27 8
54 9
65 10
48 11
25 12
14 13

Sample days: 7
Solution:

The sample mean x and sample standard deviation  x of x are:

15

h x i i
x= i =1
15
= 9.13 days
h
i =1
i

15

 h (x i i − x)2
x = i =1
15
= 2.3 days
h
i =1
i −1
 Contd..

Similarly,

For transport time

y = 9.9 days

 y =1.55 days
 Contd..

• Using a simple hypothesis test , it can be verified that the two random
variables can be assumed to be normally distributed with a mean and
standard deviation equal to sample value.

• The random variable Z associated with order cycle time is, therefore, the
sum of the independent random variable x, y .
 Contd..

• Since x, y are normally distributed, Z is also a normal random

variable with mean and standard deviation estimated by the
simple value.
z = x + y = 19.03 days

 z =  x2 +  y2 = 2.77 days
 Contd..

0.3

0.25
Probability density

0.2

0.15 X Variable
Y Variable
Z Variable
0.1

0.05

0
0 5 10 15 20 25 30

Days

Fig. 1
 Contd..


• The co-efficient of variation cv = is a ratio of standard

deviation and mean of the order cycle time.

• It can be used as a measure of the reliability of the service

offered to the customer.
Relation between order cycle time and customer inventory
• If customer is using 10 units of products per day.

• Supplier order cycle time is 8 days.

80
• The customer average inventory during order cycle time is = 40 units .
2
• If the order cycle time is reduced to 4 days.
40
• Customer average inventory during order cycle time is = 20 units .
2
 Contd..

60

50
Units of Inventory

40

30

20

10

0
0 2 4 6 8 10 12

Order Cycle (days)

Fig. 2
 Key Performance Indicator (KPI)

• Indicators are used for monitoring the logistics

activities.
• Identification of key performance indicators (KPI)
• Generally there is no valid pre defined set of KPIs
• The choice is determined by the specific logistics
system considered.
• KPIs are classified into several families,
belonging to three categories:
 Key Performance Indicator (Contd.)

➢Volume: to measure quantity of work involved in a

logistics activity belonging to this category;

➢Effectiveness: to measure quality of output

provided by a logistics activity belonging to this
category;

➢Efficiency: to measure costs generated by

functioning of a logistics activity belonging to this
category
 Key Performance Indicator (Contd.)

• Each KPI is measured according to a suitable

metric that can vary from case to case;

• To make all the measures homogeneous and for a

coherent control panel, a normalized operation for
each single indicator is needed.

• It leads to a single homogeneous scale.

 Key Performance Indicator (Contd.)

• To develop the control system by means of KPI are

summarized as follows

➢Classification of the logistics activities and

definition of the objectives to pursue;

➢Identification of the KPIs subdivided by families;

➢Creation of the control panel.

 Example of Different KPIs

Family KPI Computing Method

Dispatch of Order No of Orders Total orders received in the
month/days in the month
Complaints Number of complaints per
month
Extent of Orders dispatched overall in
Completeness the month/orders inserted
overall in the month
Errors on Number of orders with
orders delivery in the month/total
orders received in the month
 Example of Different KPIs (Contd.)

Family KPI Computing Method

Transport Deliveries per Monthly average number of customers served by
journey a route
Trip saturation Monthly average amount of goods dispatched per
vehicle/vehicle capacity
Trip forecast Trips planned in the month/effective trips in the
month

Warehouse Warehouse Monthly average of daily movements

movements
Pick-up operations Monthly number of movements/monthly number
of order lines dispatched
 Example of Different KPIs (Contd.)

Inventory Warehouse Overall monthly value of

value inventory

Costs Budget Monthly costs or monthly budget

Reliability Reliable Monthly order lines delivered

deliveries correctly on time/total monthly
order lines
 Control Panel for KPI
Steps to develop control panel

• First KPIs are normalized in the interval [1, 10] using min-
max normalization procedure on the historical data (say for
last12 data set) available for each KPI

• The normalization value equal to 1 corresponding to the worst

value (e.g. maximum no of complaints per month) whereas 10
is the best one.

• Assign weight to each member of the family suitably so that

sum of weights of every KPI family is equal to 1 so as to find
a single effective value for every family of KPI
Procedure for Normalization of Data

• Normalization of a time series yt , t = 1, 2....T is obtained by transforming it

into another time series yt' , t = 1, 2....T , where observations belong to a given
interval [min, max].

Where, min  max

• Each value yt , t = 1, 2....T will correspond to

yt − a
y =
'
(max − min) + min
b−a
t

y
• Where, a and b represent the minimum and maximum
t

value of t = 1, 2....T
 Control Panel for KPI (Contd.)

• Logisticians determine the minimum value to be

achieved for each KPI (say for example 6 and
objective should be 10)

• The control panel is constructed using Radar Graph

with greater visual and information impact

• This helps Logisticians to identify the needing

priority action
Example:
Family KPI Value Weight Family
Performance

Dispatch of Total orders received in the 4.72 0.25 5.73

Order month
Number of complaints per 6.21 0.25
month
Extent of completeness of order 5.77 0.30

Errors on number of orders with 6.34 0.20

delivery
Warehouse Warehouse movement 6.27 0.60 6.25

Pick up operation 5.96 0.40

Transport Delivery per Journey 6.34 0.5 6.78

Ratio of trip planned and 7.22 0.5

effective trip in a month
Reliability Reliable delivery 4.88 1 4.88
 Contd..

Dispatch of
Order
7

Reliability 0 Warehouse
Performance of Parent System

Transport

Fig . Control Panel (Radar Graph)