Available via license: CC BY 4.0

Content may be subject to copyright.

84

VOJNOTEHNIČKI GLASNIK/MILITARY TECHNICAL COURIER, 2013., Vol. LXI, No. 2

MEASURING EFFICIENCY

IN LOGISTICS

Milan M. Andrejić, Faculty of Transport and Traffic Engineering,

Logistics Department, University of Belgrade

Abstract:

Dynamic market and environmental changes greatly affect operating

of logistics systems. Logistics systems have to realize their activities and

processes in an efficient way. The main objective of this paper is to

analyze different aspects of efficiency measurement in logistics and to pr-

opose appropriate models of measurement. Measuring efficiency in logi-

stics is a complex process that requires consideration of all subsystems,

processes and activities as well as the impact of various financial, operati-

onal, environmental, quality and other factors. The proposed models have

a basis in the Data Envelopment Analysis method. They could help mana-

gers in decision making and corrective actions processes. The tests and

results of the model show the importance of input and output variables se-

lection.

Key words: efficiency, logistics, measuring methods, data envelopment

analysis.

Introduction

odern product and service markets are very dynamic with a

great number of participants. In order to stay in the market,

companies constantly have to measure and monitor their performances

as well as to define appropriate corrective actions. Depending on indu-

stries and types of systems, there are a large number of different perfor-

mances. One of the basic and frequently used performances is

efficiency. Efficiency is a very important indicator of the analysis of com-

panies’ operations.

There is no universal and generally accepted definition of efficiency.

Different authors define efficiency in different ways. In (Gleason, Barnum,

1982, pp. 379-386), the notion of effectiveness means the level of the

goals accomplishment ("doing the right things"), while under the notion of

efficiency they mean the accomplishment of these goals in the best pos-

sible way ("doing the right things in the right way"), i.e. accomplishing the

M

DOI: 10.5937/vojtehg61-1756

FIELD: Logistics, Operations Management

ARTICLE TYPE: Review Paper

m.andrejic@sf.bg.ac.rs

85

Andrejić, M., Measuring efficiency in logistics, pp. 84–104

largest number of outputs while using the least amount of resources. In

the past, both in literature and in practice, the greatest attention was paid

to the operational efficiency. Operational efficiency can be defined as the

ratio between the exploited resources and the accomplished results.

The efficiency measurement process in manufacturing companies is

completely different from the efficiency measurement process in service

companies. Raw materials and components in the production process

are transformed in the final product. Tangibles and easy measurability of

the final product and resource usage greatly facilitate the efficiency mea-

surement of production processes. On the other hand, the final product in

service companies is a realized service that is by its nature transient, in-

tangible and quantity immeasurable, for which realization is often

necessary to employ different measurable resources such as space, ti-

me, labor, etc.

Products of logistics companies as well as typical service companies

are: transport services, warehouse services, material-handling services,

freight forwarding services and other services. For their realization, it is

necessary to use various resources such as transport and material han-

dling equipment, warehouse space, time, energy, labor, etc. whose usa-

ge is not easily measurable. An additional problem of measuring

efficiency in logistics is the complexity, integration and mutual dependen-

ce of both resources involved and realized services.

When defining performances of logistics systems, it is possible to

make very different and even conflicting aspects of performances. Defi-

ning and measuring the efficiency of logistics systems as one of the most

important performances in recent years is very significant. Existing mo-

dels for measuring and monitoring efficiency are not fully applicable in

logistics and contain a large number of constraints. The lack of models

for measuring and monitoring the efficiency in logistics systems applica-

ble in practice is evident. This confirms the lack of papers and models

tested on real examples. The aforementioned problem and the importan-

ce of monitoring and measuring the efficiency of logistics services for

practitioners and researchers are the main motive of this paper.

Recently, an increasing number of authors have been advocating the use

of approaches such as the Data Envelopment Analysis (DEA) method

(Min, Joo, 2006, pp. 259-65). The DEA method is used for estimating the

efficiency of homogeneous Decision Making Units (DMUs).

This paper provides an overview of models for measuring the

efficiency in logistics. A special emphasis is placed on DEA models, the

most frequently used method in efficiency measurement. The aspects of

efficiency measurement in logistics are also investigated. Different mo-

dels tested on real examples are described in this paper.

86

VOJNOTEHNIČKI GLASNIK/MILITARY TECHNICAL COURIER, 2013., Vol. LXI, No. 2

Aspects of measuring efficiency in logistics

Measuring efficiency in logistics can be viewed from various

aspects. From the point of time and the significance of the decision-ma-

king, there are three hierarchical levels of efficiency measurement: stra-

tegic, tactical and operational level. In accordance with the mentioned

aspects, it is possible to make a difference among the following efficiency

measurement aspects in logistics: activities efficiency, processes effici-

ency, subsystems efficiency, systems efficiency and supply chain effici-

ency (Fig. 1).

DC

Retail

DC

DC

DC

Supply chain efficiency

Logistics system efficiency

Logistics subsystem efficiency

Logistics activities and processes efficiency

Figure 1 – Different perspectives on efficiency measuring

Slika 1 – Različite perspektive merenja efikasnosti

Measuring and monitoring the efficiency of logistic networks, chains,

systems, processes and activities primarily refers to the direct logistics

flows, but the efficiency of reverse logistics flows is also important. In ad-

dition to monitoring and measuring the efficiency in logistic systems (fre-

ight forwarding, logistic centres, distribution centres, warehouses, etc.), it

is necessary to analyze the efficiency of logistic activities and processes

in systems whose main function are not logistic services (retail chains,

hospitals, banks, etc.) For a successful measurement of the efficiency in

logistics, it is necessary to consider a large number of inputs and outputs

different in nature. Financial, technical, environmental, energy, social and

many other factors expressed in different units influence efficiency in lo-

gistics. In this sense, it is possible to measure: energy efficiency, eco

efficiency, cost efficiency, quality efficiency, etc. It can be concluded from

the previous discussion that measuring efficiency in logistics is a complex

87

Andrejić, M., Measuring efficiency in logistics, pp. 84–104

process that requires consideration of all subsystems, processes and ac-

tivities as well as the impact of various financial, operational, environ-

mental, quality and other factors.

Efficiency measuring models in logistics

From the initial works (Farrell, 1957, pp. 253-290 and Charnes, et al,

1978, pp. 429-444) and setting the basis of the DEA method as well as

from introducing the notion of a DMU (Decision Making Unit), an expan-

sion of the works from this area occurs. The suitability of application and

the quality of the obtained results have influenced the application of this

method in different profit and non-profit organizations: banks, libraries,

hospitals, schools, universities, kindergartens, etc.

In the field of logistics, the DEA method is mostly used for the evalua-

tion of the efficiency of 3PL (Third Party Logistics) providers from the per-

spective of customers and the provider (Min, Joo, 2006, pp. 259-65). The

authors in (Zhou, et al, 2008, pp. 262-279), with the help of the DEA met-

hod, attempt to define the benchmark performances variables for 3PL pro-

viders in China. In their work, they consider the change of efficiency during

a period of time as well as the interaction of certain factors upon the per-

formances. In (Hamdan, Rogers, 2008, pp. 235–244), the authors apply

the DEA method to measure the efficiency of 3PL providers with the stress

upon the warehouse operations. Some papers refer to the estimation of

distribution centres (DC) and warehouse efficiency. Thus in (Ross, Droge,

2002, pp. 19-32), the authors examine the efficiency of 102 DCs as parts

of complex supply chains as well as the change of this efficiency over the

course of time. A fuzzy DEA model for measuring the efficiency of tran-

sport quality is proposed in (Pamučar, 2011, pp. 40-61). In (Ilić, Andrejić,

2011, pp. 93-115), the authors analyse the efficiency of possible queuing

models at gas stations using the large-scale queuing theory. Benchmar-

king and monitoring the international operators of the warehouse perfor-

mances by applying the DEA method was given in (de Koster, Balk, 2008,

pp. 1-10). The method has been in many ways applied to measuring

efficiency in supply chains (de Koster, Balk, 2008, pp. 1-10).

Supply chain efficiency measurement

Supply chains are complex systems composed of many interrelated and

conditioned processes. Supply chains and similar complex systems and pro-

cesses are viewed as "black boxes" for a long time. Their structure and ope-

rating are not considered properly. For a successful efficiency evaluation of

88

VOJNOTEHNIČKI GLASNIK/MILITARY TECHNICAL COURIER, 2013., Vol. LXI, No. 2

supply chains, it is necessary to measure the performances of all participants

in the chain, including suppliers, manufacturers, traders and end users.

Some papers in the literature measure the efficiency of independent

participants in the supply chain using the DEA method. However, the

number of papers that analyse the efficiency of supply chains is very

small. The problem of measuring the efficiency of supply chains in the

literature has been recognized as a problem of measuring the efficiency

of multi-stage processes. The authors in (Cook, et al, 2010, pp. 423-430)

have categorized DEA models into four categories:

– standard DEA approach;

– efﬁciency decomposition approach;

– network DEA approach;

– game-theoretic approach.

It is important to know that each supply chain member has its own

strategy of achieving efficiency. Sometimes, due to possible conflicts be-

tween supply chain members, one member’s inefficiency may be caused

by another’s efficient operations. For example, the supplier may increase

its raw material price to enhance its revenue and to achieve an efficient

performance. This increased revenue means increased cost to the manu-

facturer. Consequently, the manufacturer may become inefficient unless

it adjusts its current operating policy. Measuring supply chain performan-

ce becomes a difficult and challenging task because of the need to deal

with the multiple performance measures related to the supply chain

members and the need to integrate and coordinate the performance of

those members (Liang, et al, 2006, 35-49).

Two hurdles are present in measuring the performance of supply cha-

ins. One is the existence of multiple measures that characterize the perfor-

mance of each member in a supply chain. The other is the existence of con-

flicts between supply chain members with respect to specific measures.

The problem of input and output variables classification is a big pro-

blem in supply chain efficiency measuring. Some measures linked to re-

lated supply chain members cannot be simply classified as “outputs” or

“inputs” of the supply chain. For example, the supplier’s revenue is not

only an output of the supplier (the supplier wishes to maximize it), but

also an input to the manufacturer (the manufacturer wishes to minimize

it). Simply minimizing the total supply chain cost or maximizing the total

supply chain revenue (profit) does not model and solve the conflicts. The-

refore, the meaning of supply chain efficiency needs to be carefully defi-

ned and studied. Models need to define and measure the efficiency of a

supply chain as well as supply chain members. A standard DEA appro-

ach uses two separate DEA runs for two supply chain members and cal-

culates independent efficiency for each member. This approach does not

treat common measure in a coordinated manner (Fig. 2).

89

Andrejić, M., Measuring efficiency in logistics, pp. 84–104

Figure 2 – Two-stage supply chain

Slika 2 – Dvoetapni lanac snabdevanja

Efficiency decomposition models define the efficiency of the overall

two-stage process. This approach takes into account the equality of in-

termediate measures. In literature, overall efficiency is defined as the

sum or the product of efficiencies of individual stages (members). Net-

work DEA models analyse a complex structure where intermediate mea-

sures are not only inputs to the second stage. These models are applied

in more general situations than two-stage processes.

The fourth type of approach uses game theory concepts. It origina-

tes from (Liang, et al, 2006, 35-49) where DEA is used to measure the

performance of supply chains with two members (as in a manufacturer–

retailer setting, for example). The concepts of the Stackelberg game (or

leader–follower) and the cooperative game are used to develop models

for measuring performance in supply chain settings. There are two main

approaches: non-cooperative and cooperative models. In non-

cooperative models, one-stage (member) is a leader while the other sta-

ge is a follower. The leader maximizes its efficiency regardless of the fol-

lower efficiency. The follower maximizes efficiency and with constraint

that leader’s efficiency is the same. The cooperative efficiency evaluation

model maximizes the joint efficiency of the both members and forces the

two players to agree on a common set of weights on the intermediate

measures. In (Andrejić, Kilibarda, 2011, pp. 237–242), a practical appli-

cation of this approach is tested.

Measuring efficiency of logistics systems

and subsystems

DCs represent complex systems with a large number of interconnec-

ted subsystems, processes and activities. They also represent the basic

hubs of the goods distribution process and connect a large number of

participants in supply chains. In order to maintain the competitive edge,

they have to monitor and measure the efficiency, but also define the sui-

Member 1

Member 2

Inputs

Intermediate measures

Outputs

90

VOJNOTEHNIČKI GLASNIK/MILITARY TECHNICAL COURIER, 2013., Vol. LXI, No. 2

table corrective actions, if necessary. The managers and employees in

such systems are faced with several different tasks that are mainly con-

nected with determining the optimal amount of the resources (input) for

realizing a certain number of shipments, of a satisfactory level of quality.

Only such systems can satisfy both the demands of the customers and

their own demands. Apart from a DC as a system, this section describes

in more detail two basic subsystems, which are the transport and ware-

house subsystems. Regardless of the strong connection, these

subsystems have different goals, in different cases even conflicts. Just as

DCs, these subsystems strive to maximize the efficiency. In this sense,

DC managers must at the same time take into account the efficiency of a

DC as a system together with its subsystems.

Model formulation for logistics systems

and subsystems efficiency measurement

In the process of applying the DEA and Multiple Objective Data Envel-

opment Analysis (MODEA) approaches, one of the most important steps re-

presents the selection of the input and output variables (Boussofiane, et al,

1991, pp. 1-15). This work has evaluated the efficiency of 20 DCs of three

trading companies who operate in the region of Serbia and who have similar

sale networks, products ranges and distribution systems (Kilibarda, et al,

2011, pp. 996-1010). In order to estimate the DC and warehouse and tran-

sport subsystem efficiency, the variables describing their functioning in the

best way have been used. The number of employees, the number of pallet

places and the number of distribution vehicles at the DC disposal represent

the main input variables, while the number of realized deliveries, the number

of errors in the warehouse subsystem and the number of errors in the tran-

sport subsystem represent the output variables used in this work. The num-

ber of employees in a DC represents the common input variable, while the

number of realized deliveries represents the common output variable in the

transport and warehouse subsystem. It is assumed that DCs realize the deli-

veries the sizes of which are approximately the same. As the second input

variable to estimate the warehouse subsystem efficiency, the number of pal-

let places in the DC has been used. On the other hand, the number of vehic-

les at DC disposal represents an input variable in the transport subsystem.

The mentioned variables primarily refer to the estimation of the operational

efficiency. In order to obtain the results whose quality and reliability are

higher, as additional output variables, the quality indicators have been used.

DCs of trading companies, but also DCs in general are characterized by a

large number of indicators of realized deliveries. Errors in the transport and

warehouse subsystems represent qualitative indicators which may be the

91

Andrejić, M., Measuring efficiency in logistics, pp. 84–104

cause of dissatisfaction and complaints on the part of the customer, i.e. a low

service quality level. Errors in the warehouse subsystem are mostly related to

the errors in the order-picking process (shortage/excess in the delivery, arti-

cles mix-up, damage), but also to other processes such as bad inventory

management, etc. Errors in transport primarily concern the delivery that is

falling behind schedule, as well as the damaging and losing goods in the

transport process. As it can be seen in Fig. 3, in order to estimate the DC ef-

ficiency, six variables have been used (three input and three output vari-

ables), while, in order to estimate the efficiency of its subsystems, four vari-

ables have been used per each system (two input and two output ones).

Figure 3 – DC and basic subsystems

Slika 3 – DC i osnovni podsistemi

The variables of input and output variables are given in Table 1. Table

1 also offers the descriptive statistics of the variables used in this paper. In

order to obtain the results of higher quality and reliability, on the recommen-

dation in (Klimberg, et al, 2010, pp. 79-93), the normalization of all variables

has been performed. The number of errors in the transport and warehouse

subsystems represents the output variables of the negative orientation, i.e.

smaller variables are favorable. Regarding the fact that the DEA method

perceives all output variables as positively oriented (larger variables are fa-

vorable), the change in the orientation of the mentioned variables has been

performed. Prior to defining the model to estimate the DC and its

subsystems efficiency, it is necessary to mention that the model proposed in

this paper does not consider the external factors which may influence their

efficiency, and do not come within the competence of the management of

the company. The mentioned factors may be: weather conditions, market

situation, industry branch, competition behavior, etc.

Number of

pallete places

Employees

Vehichles

Warehouse

flaws

Transport

flaws

Realized

Services

WAREHOUSE

SUBSYITEM

TRANSPORT

SUBSYSTEM

DISTRIBUTION

CENTER

92

VOJNOTEHNIČKI GLASNIK/MILITARY TECHNICAL COURIER, 2013., Vol. LXI, No. 2

Table 1

Inputs and output variables in DC

Tabela 1

Ulazne i izlazne promenljive distributivnog centra

DMU Employees

Palette

places

Vehicles

Realized

deliveries

Warehouse

errors

Transport

errors

Mean 98.26 3941.10 27.96 4544.95 483.65 749.25

Standard 56.10 1898.57 10.45 2730.54 375.48 444.78

Max 186 6793 45 9006 1336 1426

Min 34 684 13 1207 36 133

In this particular case, a DC (with warehouse and transport

subsystems) represents a DMU whose efficiency is estimated. This work

estimates the efficiency of 20 DCs and their transport and warehouse

subsystems by means of applying the DEA and MODEA approach. The

DEA method is a mathematical programming technique which gives an

opportunity to compare different DMUs based on multiple inputs and out-

puts. This paper utilizes the DEA approach in order to estimate the

efficiency of DCs and warehouse and transport subsystems. According to

this approach, a logistic system or subsystem represents a "black box",

i.e. the processes and activities that are not realized within the

subsystem are not taken into consideration. From the group of DEA mo-

dels, the basic CCR model has been chosen (Charnes, et al, pp. 429-

444). By applying the CCR models, an independent estimation of the DC,

warehouse and transport subsystems has been performed.

The described model belongs to the group of standard approaches

which perform an independent estimation of the system efficiency without

entering into the structure of the very system. In this manner, it is not pos-

sible to measure the DC efficiency taking into account the efficiency of the

transport and warehouse subsystems, regarding the fact that these two

subsystems possess common input and output variables. The CCR DEA

models are not "sensitive" enough when evaluating the DMU efficiency,

regarding the fact that they do not take into consideration the compromises

and conflicts of independent DC subsystems goals. The MODEA approach

offers the possibility of overcoming the mentioned problem (Klimberg,

Puddicombe, 1999, pp. 201-232). The MODEA does not represent a sim-

ple set of independent standard CCR DEA models, but takes into account

the common resources and multiple goals of the subsystems. In the model

suggested in this section, the common input variable is the number of

employees, while the common output variable is the number of realized

deliveries. In most of the cases, the contribution of the common variable in

one subsystem depends upon the contribution of the same variable in the

other subsystem. In the ideal case, the ratio of the common variables sho-

93

Andrejić, M., Measuring efficiency in logistics, pp. 84–104

uld be approximately 1 (the contribution of the common variable is identi-

cal). By further limitations upon the influence of the common variables in

different goals on the part of the decision maker, the model becomes "clo-

ser" to the real state of affairs and good operational practice. The examina-

tion of the weights of the variables and their influence within the goals

enables the decision makers to reach a suitable compromise. In the

analyzed example, the set of decision making units makes 20 DC (DMU)

(k = 1, 2, ..., 20), with two subsystems each (p=1,2). The warehouse and

transport subsystems are characterized by i input variables (i=1,2, ..., m

p

)

and r output variables (r=1,2, ..., s

p

). In this particular case, the transport

and warehouse subsystems have two input and output variables each

(m

1

= m

2

= s

1

= s

2

=2). The variable of the r-th output of the p-th subsystem of

the k-th DC is marked by y

rkp

, while x

ikp

marks the variable of the i-th output

of the p-th subsystem of the k-th DC. The weight coefficient assigned to

the r-th output of the p-th subsystem u

rp

and the weight coefficient assig-

ned to the i-th output of the p-th subsystem v

ip

represent the unknown vari-

ables (decision making variables). The coefficient

ϕ

represents the para-

meter of closeness of the common resources, assigned by the decision

maker. In the concrete case, the MODEA model for measuring the

efficiency of the k-th DC has the following form:

221121

2

1

rkrrkrkk

p

kp

yuyuwww Max +=+=

∑

=

(1)

1,2p xv

p

m

i

ikpip

==

∑

=

,1

1

(2)

∑∑

==

=∀=∀≤−

pp

m

i

ikpip

s

r

rkprp

k1,2;pxvyu

11

20210 ,...,,, (3)

ϕϕ

+≤≤− 11

2

2

1

1

s

u

s

u

r

r

(4)

ϕϕ

+≤≤− 11

2

2

1

1

m

v

m

v

i

i

(5)

210 ,, =

=

≥ i1,2;pv

ip

(6)

210 ,, =

=

≥ r1,2;pu

rp

(7)

94

VOJNOTEHNIČKI GLASNIK/MILITARY TECHNICAL COURIER, 2013., Vol. LXI, No. 2

In this paper, the efficiency of 20 DCs and their transport and ware-

house subsystems has been analyzed. The set of the observed DCs can

be deemed homogenous due to the fact these DCs operate in the same

way and under the same conditions. The efficiency estimation has been

carried out by applying the CCR DEA and MODEA approaches. The first

three models refer to the MODEA approach, in which the coefficient φ

(closeness parameter) takes the variables of 10%, 20% and 50%. By

comparing the number of DC efficiencies obtained by applying the MO-

DEA and CCR DEA approaches, it is pointed to a considerably larger

number of efficient DMUs by applying the latter approach. This can be

explained by the fact that the CCR model joins all indicators into a unique

measure of efficiency with no separation or observation of the efficiency

of its subsystems, i.e. it "overrates" the efficiency of DC.

By applying the MODEA approach, only 15% efficient DC can be ob-

tained, while the CCR approach yields even 45% efficient DC. Such re-

sults can be explained by the fact that the CCR model does not take into

consideration the efficiency of subsystems in a DC. On the other hand,

according to the MODEA approach, a DC is efficient if the transport and

warehouse subsystems are efficient. The average DC efficiency by

applying the MODEA approach amounts to 1.1388 with the average

efficiency of the warehouse subsystem amounting to 0.5158, and the

transport one to 0.6229. The observed DC set can be considered as

relatively inefficient and the largest portion of this inefficiency is a conse-

quence of the warehouse subsystem inefficiency (Kilibarda, et al, 2011,

pp. 996-1010).

The efficiency of one DC subsystem (warehouse or transport one)

does not entail the DC efficiency. It can easily be concluded that the wa-

rehouse subsystem efficiencies are smaller than the transport subsystem

efficiencies regardless of the variable of the parameter φ. By analyzing

the obtained variables, it can be concluded that there is a certain number

of DCs whose subsystems do not change the efficiency regardless of the

approach. DMU 2, DMU 12 and DMU 17 represent the DCs with stable

performances and efficient according to both the MODEA and the CCR

DEA approach. Stable efficiency of these centres can be explained by a

relatively small quantity of the used resources for the realization of a gre-

at number of deliveries with the number of errors in the transport and wa-

rehouse subsystems DMU 2 and DMU 12 considerably lower than the

average (Kilibarda, et al, 2011, pp. 996-1010).

By analyzing Fig. 4 in more detail, three characteristic DC groups

can be selected. The first group is comprised of the formerly mentioned

efficient units. This group includes DMU 2, DMU 12 and DMU 17 whose

transport and warehouse subsystems are efficient. The second group

includes the units whose warehouse subsystems are significantly more

95

Andrejić, M., Measuring efficiency in logistics, pp. 84–104

efficient than the transport subsystems. Such units are DMU 1 and DMU

11. As opposed to them, DMU 3, DMU 6, DMU 8, DMU 9 and DMU 13

have significantly more efficient transport subsystems. The last two DC

groups, apart from inefficiency of at least one subsystem are characteri-

zed by the overall DC inefficiency. This confirms the claim that no DC

can be efficient if one of its subsystems is inefficient.

Figure 4 – Efficiency of a DC and its subsystems

Slika 4 – Efikasnost DC i osnovnih podsistema

Measuring efficiency of refrigerated warehouses

Energy efficient warehouses and distribution centers are integral

components of supply chain strategies, especially nowadays in competi-

tive global economy. Warehouses are moving beyond a simple storage

function and today they offer high-tech inventory tracking and value-

Distributivni centar

0

0.5

1

1.5

2

2.5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

DMU

Efikasnost

MODEA 10%

MODEA 20%

MODEA 50%

Skladišni podsistem

0

0.2

0.4

0.6

0.8

1

1.2

1234567891011121314151617181920

DMU

Efikasnost

MODEA 10%

MODEA 20%

MODEA 50%

Single objective

Distributivni centar

0

0.5

1

1.5

2

2.5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

DMU

Efikasnost

MODEA 10%

MODEA 20%

MODEA 50%

Transportni podsistem

0

0.2

0.4

0.6

0.8

1

1.2

1 2 3 4 5 6 7 8 9 1011121314151617181920

DMU

Efikasnost

MODEA 10%

MODEA 20%

MODEA 50%

Single objective

a) DC Efficiency

b) Warehouse efficiency

c) Transport efficiency

Efficiency

Efficiency

Efficiency

CCR

CCR

96

VOJNOTEHNIČKI GLASNIK/MILITARY TECHNICAL COURIER, 2013., Vol. LXI, No. 2

added services as quality-control testing and repackaging. As warehou-

ses grow more sophisticated, their energy consumption grows too. For

example, in the U.S., warehouses only for lighting spend an average of

10$/m

2

. Energy costs for some warehouses are more than 10% of their

total revenue. It does not matter whether a production, warehouse or di-

stribution center is 1000 square meters or 500000 square meters, it is

necessary to cut costs and to improve operating efficiencies. Refrigera-

ted warehouses (RWs) represent special types of warehouses that use

2.8 times more energy for operating than conventional warehouses (Roy,

2010, pp. 64-70). RWs are very important links in a supply chain (AS-

HRAE, 2006). They are also large energy consumers.

Logistic systems, especially RWs, demand more energy for process ope-

ration. In a situation of increasing energy demand and rising energy costs,

energy efficiency measures are becoming more important. The need for

energy efficiency measuring and monitoring becomes a necessity. This paper

uses one of the most frequently used methods for measuring efficiency – the

DEA method for RWs energy efficiency measuring. According to this method,

RWs are viewed as systems that use inputs (labor, electricity, fuel, water,

space, etc.) to produce products and deliver services with a high level of qua-

lity. Management plays the key role in the energy consumption reduction. This

paper presents a model for evaluating RWs energy efficiency, which should

help managers to reduce energy consumption.

The constant return to scale (CRS) assumption is appropriate when all

firms operate on an optimal scale. The CCR gives relative technical

efficiency (TE). However, imperfect competition, government regulations,

constraints on finance, etc., may cause a firm not to operate on an optimal

scale. In contrast to the CCR, the model that assumes CRS (proportional

increase in inputs results in a proportional increase in outputs), (Coelli, Pra-

sada, O'Donnell, Battese, 2005) extends the original DEA model to variable

returns to scale (VRS - increase in inputs does not result in a proportional

change in output). This model is known in the literature as the BCC model.

The BCC model measures the pure technical efficiency (PTE), and provides

a measure of efficiency which ignores the influence of the volume of busi-

ness, because RWs are compared only with other RWs of a similar scale.

Scale efficiency (SE) indicates whether the observed RW operates with an

optimal volume of operations. SE can be obtained for each RW by conduc-

ting both CCR and BCC models, and then by decomposing the TE scores

obtained from the CCR model into two components, one due to scale

inefficiency and one due to pure technical inefficiency. If there is a difference

in CCR and BCC TE scores for a particular RW, then this indicates that the

firm has scale inefficiency (Coelli, Prasada, O'Donnell, Battese, 2005):

TE (Technical Efficiency - CCR) = PTE (Pure Technical Efficiency -

BCC) x SE (Scale Efficiency)

97

Andrejić, M., Measuring efficiency in logistics, pp. 84–104

It is important to note that the overall efﬁciency of energy consump-

tion consists of two factors: environmental factors and energy manage-

ment factors. The environmental factors include weather conditions, oc-

cupant intensity, imperfect competition, government regulations, constra-

ints on finance, etc. The energy management factors include equipment

efﬁciency, operating strategy, etc. Due to the difﬁculty of separating the

effect of the environmental factors from overall energy efﬁciency, there is

little research analysis of the effect of energy management of RWs with

the DEA. Therefore, we use the DEA to measure the overall energy

efﬁciency in detail by examining the environmental factors and the ma-

nagement factors. Considering the fact that RWs can control energy con-

sumption and reduce it in order to become more efficient, in this paper,

we adopted the input orientation of DEA models.

Five inputs and two outputs are used for evaluating the efficiency of

15 RWs. The data are generated in accordance with the actual values of

RWs in real systems (Andrejić, et al, 2011, pp. 313-319). The following

inputs are taken into consideration (Table 2): RW size (m

3

), labor, facility

utilization (%), electricity consumption, other energy costs (water, fuel,

gas, etc). The RW size presents the facility volume. It is better to use m

3

rather than m

2

when describing the warehouse space. The percentage of

facility utilization is a good indicator for the rationalization of energy con-

sumption. Electricity consumption is a basic measure for energy efficient

measuring. As mentioned before, electricity costs are more than 50% of

all costs in RWs, so a detailed analysis of this input is necessary. This

measure is expressed in MWh/year. All other costs of water, fuel, gas,

etc., consumption are incorporated in one measure – other energy costs,

and they are expressed in thousands of monetary units (m.u.).

On the other hand, the proposed model uses two outputs:

successfully realized services and goods deterioration. Successfully rea-

lized services are services that are realized according to customer requi-

rements (right time, right place, right quality, etc). Goods deterioration is

caused by poor handling and storing of goods (i.e. inappropriate tempe-

rature, pressure and humidity).

In accordance with the defined input and output values, as mentio-

ned, we used the input minimizing orientation of CCR and BCC models,

where the emphasis is on the maintenance or improvement of the level of

service provided with the minimum use of inputs. In this paper, the

analysis is carried out with the DEA software efficiency measurement

system (EMS), developed by the Operations Research Department at the

University of Dortmund ((http://www.ub.uni-dortmund.de/service/benu-

ord.htm.en). Table 2 presents the descriptive statistics for the outputs

and inputs of the 15 RWs.

98

VOJNOTEHNIČKI GLASNIK/MILITARY TECHNICAL COURIER, 2013., Vol. LXI, No. 2

Table 2

Descriptive statistics for the inputs and outputs in refrigerated warehouses

Tabela 2

Deskriptivna statistika ulaznih i izlaznih promenljivih skladišta hladnjača

Inputs/outputs Mean

Standard

deviation

Max Min

Size (m3) 51.67 23.29 92.72 26.36

Employees 74.68 29.72 132.69 31.95

Facility utilization (%) 69.20 17.64 98.20 44.48

Electricity consumption (MWh/year) 3993.44 843.75 5078.87 2420.65

Other energy costs (1000 m.u.) 26.66 11.04 48.01 11.73

Realized services (1000) 535.79 215.94 997.13 298.57

Goods deterioration (%) 9.54 5.89 19.37 1.12

The results of the DEA analysis, TE and SE scores for each RW are

given in Table 3. Out of the 15 RWs, 8 (53%) were found TE with a score

of 100%. The remaining 7 RWs (47%) were technically inefficient since

they had a TE score lower than 100%. The average efficiency of 15

analyzed RWs is 85%, which roughly means that each RW can store and

deliver the current quantities of goods up to 15% less resources during the

observed period (in this case one year). In this way, the RWs can realize

potential savings of 600 MWh in electricity consumption and 4000 m.u. in

other energy costs in average. In average, the RWs could operate with 11

employees less. The average technical inefficiency of the observed RWs is

15%. Most of the inefficiency is in SE, about 14%, while about 1% of

inefficiency corresponds to PTE. These results show the fact that most of

the inefficiency is the result of the inadequate size of the facility, and the

environmental factor rather than management (i.e. an RW does not opera-

te on an optimal scale). In real systems, the inefficiency is often caused by

bad management. The detailed analysis of the obtained results shows that

the RWs can be classified into certain groups. Namely, the DEA analyses

have found that there are 8 technically efficient and 7 inefficient RWs. As

mentioned before, most RWs do not operate on the optimal scale. A rea-

son for their inefficiency can be found in the size of RWs and the environ-

mental factors (Andrejić, et al, 2011, pp. 313-319).

Table 3

RWs efficiency scores

Tabela 3

Efikasnost skladišta hladnjača

RW TE (CRS) PTE (VRS) SE

RW 1 0.54 1.00 0.54

RW 2 1.00 1.00 1.00

RW 3 1.00 1.00 1.00

99

Andrejić, M., Measuring efficiency in logistics, pp. 84–104

RW TE (CRS) PTE (VRS) SE

RW 4 0.53 0.98 0.54

RW 5 1.00 1.00 1.00

RW 6 1.00 1.00 1.00

RW 7 0.80 1.00 0.80

RW 8 1.00 1.00 1.00

RW 9 1.00 1.00 1.00

RW 10 0.61 1.00 0.61

RW 11 0.97 0.98 0.99

RW 12 1.00 1.00 1.00

RW 13 0.50 0.91 0.55

RW 14 0.86 1.00 0.86

RW 15 1,00 1,00 1,00

Average efficiency 0.85 0.99 0.86

Efficient units 8 (53%) 12 (80%) 8 (53%)

Inefficient units 7 (47%) 3 (20%) 7 (47%)

The slack movements,which arise because of the sections of the pi-

ece-wise linear frontier that runs parallelly to the axes are reported in or-

der to give an accurate indication of RWs TE. The analysis of slack valu-

es (Table 4) shows that there is a significant slack in the input variable.

Much of the inefficiency is the result of scale inefficiency of the RW ope-

rating. Potential improvements are related to the average decrease in

inputs, which range from 0.7% for other energy costs up to 11.21% for

employees while maintaining RW current level of outputs.

Table 4

Potential savings in inputs

Tabela 4

Potencijalne uštede u ulazima

Actual average

Average

slack

Potential

reduce

Size 51.67 3.51 6.79%

Employees 74.68 8.37 11.21%

Facility utilization 69.20 5.39 7.78%

Electricity consumption 3993.44 254.85 6.38%

Other energy costs 26.66 0.19 0.70%

Namely, RW energy efficiency heavily depends on the organization

management, technologies, size of facility and awareness of employees

in RWs, as well as of a number of external factors.

100

VOJNOTEHNIČKI GLASNIK/MILITARY TECHNICAL COURIER, 2013., Vol. LXI, No. 2

Conclusions

Measuring and monitoring the efficiency in logistics is one of the cru-

cial success factors in the market. The main objective of this paper is to

define some kind of efficiency measurement models review. This paper

describes different aspects of measuring efficiency in logistics. The main

efficiency measurement aspects in logistics described in this paper are:

activities efficiency, processes efficiency, subsystems efficiency, systems

efficiency and supply chain efficiency. The paper further describes the

fundamental issues and models for measuring efficiency in certain areas.

The tests and results of the model show remarkable importance of input

and output variables selection. In this paper, the models for measuring

supply chain, systems, and subsystems are described in more detail.

They are also tested on real examples. The methodology for measuring

refrigerated warehouses is also described in this paper.

In literature, there is a lack of case studies, i.e.model testing in con-

crete examples. This fact indicates the insufficient amount of research in

this area. The efficiency of supply chains and logistci systems is also in-

fluenced by a great number of factors upon which the company mana-

gement have no influence. In future models, it would be desirable to in-

troduce certain indicators which, to a certain extent, can describe exter-

nal factors such as: weather conditions, market situation, competition be-

havior, etc. The presented models are a good basis for the development

of future models. Incorporating different performance indicators (quality,

cost, environmental, social, etc.) in efficiency measurement models sho-

uld be a subject of future research.

References

American Society for Heating Refrigeration and Air-Conditioning Engineers

ASHRAE, 2006, Handbook: Refrigeration., I-P Edition,

Andrejić, M., Kilibarda, M., 2011, Efikasnost logističkih procesa distribucije

proizvoda, National Conference on Quality, Kragujevac, A 237-A 242,

Andrejić, M., Ratković, B., Kilibarda, M., Bjelić, N., 2011, Measuring energy

efficiency of refrigerated warehouses, React Conference, Belgrade, pp. 313-319.

Boussofiane, A., Dyson, R. G. and Thanassoulis, E., 1991, Applied Data

Envelopment Analysis, European Journal of Operational Research, Vol. 52, No.

1, pp. 1-15.

Charnes, A., Cooper, W.W., Rhodes, E., 1978, Measuring efficiency of de-

cision making units, Europen Journal of Operations Research, Vol. 2, No. 6, pp.

429-444.

Coelli, T. D., Prasada, D.S.R.., O'Donnell, C. J. and Battese, G.E., 2005,

An introduction to efficiency and productivity analysis, Springer 2

nd

ed, New

York,

101

Andrejić, M., Measuring efficiency in logistics, pp. 84–104

Cook, W.D., Liang, L. and Zhu, J., 2010, Measuring performance of two-

stage network structures by DEA: A review and future perspective, Omega, Vol.

38, No. 6, pp. 423-430.

de Koster, M.B.M. and Balk., B.M., 2008, Benchmarking and monitoring in-

ternational warehouse operations in Europe, Production and Operations Mana-

gement, Vol. 17, No. 2, pp. 1-10.

Farrell, M. J., 1957, The Measurement of productive Efficiency, Journal of

the Royal Statistical Society, Vol. 120, No. 3, pp. 253-290.

Gleason, J.M., Barnum, D.T., 1982, Toward valid measures of public sector

productivity: performance measures in urban transit, Management Science, Vol.

28, No. 4, pp. 379-386.

Hamdan, A., Rogers, K.J., 2008, Evaluating the efficiency of 3PL logistics ope-

rations, International Journal Production Economics, Vol. 113, No.1, pp. 235-244.

Ilić, S., Andrejić, M., 2011, Analiza efikasnosti mogućih modela opsluživa-

nja na pumpnim stanicama primenom teorije masovnog opsluživanja/Analysis of

effectiveness of possible queuing models at gas stations using the large-scale

queuing theory, Vojnotehnički glasnik/Military Technical Courier, Vol. 59, No. 4,

pp. 93-115, Ministry of Defence of the Republic of Serbia, Belgrade,

Kilibarda, M., Andrejić, M., Vidović, M., 2011, Measuring efficiency of logi-

stics processes in distribution centers, 14 th QMOD Conference on Quality and

Service Sciences 2011 - From LearnAbility & InnovAbility to SustainAbility, San

Sebastian. Spain, pp. 996-1010.

Klimberg, R.K. and Puddicombe, M., 1999, A Multiple Objective Approach

to Data Envelopment Analysis, Advances in Mathematical Programming and

Financial Planning, Vol. 5, pp. 201-232.

Klimberg, R.K., Lawrence, K. D. and Lawrence, S. M., 2010, Data enve-

lopment analysis is not multiobjective analysis, Applications of Management

Science, Vol. 14, pp. 79-93.

Liang, L., Yang, F., Cook, W.D., Zhu, J., 2006, DEA models for supply chain

efﬁciency evaluation, Annals of Operations Research, Vol. 145, No. 1, pp. 35-49.

Min, H., Joo, S., 2006, Benchmarking the operational efficiency of third

party logistics providers using data envelopment analysis, Supply Chain Mana-

gement: An International Journal, Vol. 11, No. 3, pp. 259-65.

Pamučar, D., 2011, Fuzzy-DEA model za merenje efikasnosti kvaiteta tran-

sporta/Fuzzy-DEA model for measuring the efficiency of transport quality, Vojno-

tehnički glasnik/Military Technical Courier, Vol. 59, No. 4, pp. 40-61, Ministry of

Defence of the Republic of Serbia, Belgrade.

Ross, A., Droge, C., 2002, An integrated benchmarking approach to distri-

bution center performance using DEA modeling, Journal of Operations Mana-

gement, Vol. 20, No. 1, pp. 19-32.

Roy, M., 2010, Green Warehouses, ASHRAE Journal, Vol. 52, No. 3, pp. 64-70.

Zhou, G., Min, H., Xu, C., Cao, Z., 2008, Evaluating the comparative efficiency of

Chinese third-party logistics providers using data envelopment analysis, International

Journal of Physical Distribution & Logistics Management, Vol. 38, No. 4, pp. 262-279.

http://www.ub.uni-dortmund.de/service/benu-ord.htm.en

102

VOJNOTEHNIČKI GLASNIK/MILITARY TECHNICAL COURIER, 2013., Vol. LXI, No. 2

MERENJE EFIKASNOSTI U LOGISTICI

OBLAST: logistika, operativni menadžment

VRSTA ČLANKA: pregledni članak

Sažetak:

Dinamično tržište i promene u okruženju u velikoj meri utiču na funkci-

onisanje logističkih sistema. Logistički sistemi moraju realizovati aktivnosti i

procese na efikasan način. U radu se analiziraju različiti aspekti merenja efi-

kasnosti u logistici i predlažu odgovarajući modeli merenja. Merenje efika-

snosti u logistici predstavlja kompleksan proces koji zahteva sagledavanje

svih podsistema, procesa i aktivnosti, kao i uticaja različitih finansijskih, ope-

rativnih, ekoloških, kvalitativnih i drugih faktora. Predloženi modeli bazirani

su na Data Envelopment Analysis metodi. Oni mogu pomoći menadžerima

u procesu odlučivanja i sprovođenja korektivnih akcija. Analizom rezultata

testiranih modela potvrđen je značaj izbora ulaznih i izlaznih veličina.

Uvod

Savremeno poslovanje prvenstveno podrazumeva izrazito zahtevnu

tržišnu borbu, bez obzira na to da li se radi o proizvodnji ili pružanju usluga

u bilo kojoj grani industrije. Postupak merenja efikasnosti u proizvodnim

kompanijama u velikoj meri se razlikuje od postupka merenja efikasnosti u

uslužnim kompanijama. Za proizvodnju materijalnog proizvoda koriste se

materijali, sirovine i komponente od kojih u proizvodnom procesu nastaje

finalni proizvod. Opipljivost i laka merljivost finalnog proizvoda, kao i upo-

trebljenih resursa, u velikoj meri olakšavaju merenje efikasnosti proizvod-

nih procesa. Nasuprot tome, kod uslužnih kompanija finalni proizvod je re-

alizovana usluga koja je po svojoj prirodi prolaznog karaktera, neopipljiva i

kvantitativno nemerljiva, za čiju upotrebu se angažuju različiti, često teško

merljivi resursi, kao što su prostor, vreme i angažovana radna snaga. Pro-

izvod logističkih kompanija, kao tipičnih uslužnih kompanija, najčešće su

transportne, skladišne, pretovarno-manipulativne, špediterske i druge

usluge, u čijoj realizaciji se angažuju različiti resursi, kao što su transport-

na i pretovarno- manipulativna sredstva, skladišni prostor, vreme, radna

snaga, itd. čiju upotrebu nije jednostavno izmeriti. Dodatni problem mere-

nja efikasnosti u logistici predstavlja kompleksnost, integrisanost i među-

sobna uslovljenost, kako angažovanih resursa, tako i realizovanih usluga.

Aspekti merenja efikasnosti u logistici

Sa vremenskog aspekta razlikuje se merenje efikasnosti na strate-

škom, taktičkom i operativnom nivou. Sa aspekta nivoa merenja efikasnosti

moguće je razlikovati: efikasnost aktivnosti, efikasnost procesa, efikasnost

podsistema, efikasnost sistema i efikasnost lanca. Kada je reč o merenju i

praćenju efikasnosti logističkih mreža, lanaca, sistema, procesa i aktivnosti

moguć

e je razlikovati efikasnost direktnih i povratnih logističkih tokova. Osim

praćenja i merenja efikasnosti u logističkim sistemima (špedicije, logistički

centri, distributivni centri, skladišta itd...) neophodno je analizirati efikasnost

103

Andrejić, M., Measuring efficiency in logistics, pp. 84–104

logističkih aktivnosti i procesa u sistemima čija osnovna funkcija nije pruža-

nje logističkih usluga (trgovinski lanci, zdravstvene ustanove itd.). Za uspe-

šno merenje efikasnosti u logistici neophodno je sagledati veliki broj ulaza i

izlaza koji su po svojoj prirodi raznorodni (finansijski, tehnički, ekološki, ener-

getski, socijalni itd.) i izražavaju se u različitim mernim jedinicama. U tom

smislu moguće je meriti energetsku, ekološku, troškovnu i druge tipove efi-

kasnosti u logistici. Sa druge strane, često je neophodno sagledati pomenu-

te aspekte i definisati jedinstvenu meru efikasnosti.

Modeli merenja efikasnosti u logistici

Za procenu efikasnosti distributivnih centara (DC) dugo su korišće-

ni „single ratio” pokazatelji, kao što su iskorišćenost opreme i produktiv-

nost radne snage. S obzirom na to da pomenuti izmeritelji nisu najbolji

pokazatelji efikasnosti, u (Min, Joo, 2006, pp. 259-65) ističu da se veliki

broj autora zalaže za upotrebu Data Envelopment Analysis (DEA) meto-

de. DEA metoda predstavlja jedan od najšire korišćenih alata za mere-

nje efikasnosti. U oblasti logistike DEA metoda se najčešće koristi za pr-

ocenu efikasnosti 3PL (Third Party Logistics) provajdera, kako iz per-

spektive provajdera (Min, Joo, 2006, pp. 259-65), tako i iz perspektive

korisnika. U (Zhou, et al, 2008, pp. 262-279), uz pomoć DEA metode,

pokušavaju da definišu benčmark vrednosti performansi za 3PL provaj-

dere u Kini. Manji broj radova odnosi se na procenu efikasnosti DC i

skladišta. Tako u (Ross, Droge, 2002, pp. 19-32) istražuju efikasnost

102 DC, kao delova kompleksnih lanaca snabdevanja i promenu te efi-

kasnosti u vremenu. Metoda je višestruko primenjivana za merenje efi-

kasnosti u lancima snabdevanja (Cook, et al, 2010, pp. 423-430).

Merenje efikasnosti u lancima snabdevanja

Iz perspektive DEA metode, kao najčešće korišćene metode za

merenje efikasnosti višefaznih procesa, postoji veliki broj modela i meto-

da koje se direktno ili uz izvesne modifikacije mogu primeniti na lance

snabdevanja. U ovom radu izrađen je pregled DEA modela za merenje

efikasnosti lanaca snabdevanja. Prema (Cook, et al, 2010, pp. 423-430)

pomenuti modeli mogu se klasifikovati u četiri grupe:

• standardni DEA modeli;

• modeli dekompozicije efikasnosti;

• mrežni DEA modeli;

• modeli bazirani na teoriji igara.

Merenje efikasnosti logističkih sistema i podsistema

Distributivni centri predstavljaju kompleksne sisteme sa velikim brojem

međusobno povezanih podsistema, procesa i aktivnosti. Radi održavanja

konkurentnosti, oni moraju pratiti i meriti efikasnost, ali i definisati odgovara-

juće korektivne akcije, ukoliko je to potrebno. Pred menadžerima i zaposle-

nima u ovakvim sistemima postavlja se više različitih zadataka koji su uglav-

nom vezani za odre

đivanje optimalne količine resursa (ulaza) za realizova-

nje određenog broja isporuka, odgovarajućeg nivoa kvaliteta. Pored DC kao

sistema u radu su detaljnije analizirana i dva osnovna podsistema i to tran-

104

VOJNOTEHNIČKI GLASNIK/MILITARY TECHNICAL COURIER, 2013., Vol. LXI, No. 2

sportni i skladišni podsistem. Bez obzira na jaku povezanost ovi podsistemi

imaju različite ciljeve, u određenim slučajevima i konfliktne. Kao i DC i podsi-

stemi teže maksimizaciji efikasnosti. U tom smislu menadžeri DC moraju

istovremeno voditi računa o efikasnosti DC, kao sistema, ali i njegovih pod-

sistema. U radu je analizirana efikasnost 20 DC i njegovih transportnih i

skladišnih podsistema. Skup posmatranih DC može se smatrati homogenim

s obzirom na to da DC posluju na isti način i pod istim uslovima. Procena

efikasnosti sprovedena je primenom CCR DEA i Multiple Objective Data

Envelopment Analysis (MODEA) pristupa. Primenom MODEA pristupa do-

bija se svega 15% efikasnih DC, dok se CCR pristupom dobija čak 45% efi-

kasnih DC. Ovakvi rezultati mogu se objasniti činjenicom da CCR model ne

razmatra efikasnost podsistema u DC. Sa druge strane, prema MODEA pri-

stupu (Kilibarda, et al, 2011, pp. 996-1010) DC je efikasan ako su i tran-

sportni i skladišni podsistem efikasni.

Merenje efikasnosti skladišta hladnjača

Energetski efikasna skladišta i distributivni centri sastavni su delovi

strategije lanca snabdevanja, posebno danas u konkurentnoj globalnoj

ekonomiji. U situaciji povećanja potrošnje energije i rasta troškova ener-

gije, energetska efikasnost postaje sve važnija. Potreba za merenjem i

praćenjem energetske efikasnosti postaje nužnost.

Pet ulaznih i dve izlazne veličine korišćene su za procenu efikasnosti

15 skladišta hladnjača. Ulazne veličine korišćene u ovom modelu su: veličina

objekta, iskorišćenost objekta, potrošnja energije, ostali energetski troškovi

(naftni derivati, gas, voda itd.). Izlazne veličine, razmatrane u predloženom

modelu, jesu: uspešno realizovane usluge i stepen kvarenja robe. Analizom

rezultata zaključeno je da efikasnost skladišta hladnjača u velikoj meri zavisi

od menadžmenta, tehnološke razvijenosti i odgovornosti zaposlenih, a mno-

go manje od spoljnih faktora (Andrejić, et al, 2011, pp. 313-319).

Zaključak

Merenje i praćenje efikasnosti u logistici jedan je od presudnih fak-

tora uspeha na tržištu. Osnovni cilj ovog rada je pregled modela merenja

efikasnosti. U radu se opisuju različiti aspekti merenja efikasnosti u logi-

stici,, od kojih su osnovni: efikasnost aktivnosti, efikasnost procesa, efi-

kasnost podsistema, efikasnost sistema i efikasnost lanca snabdevanja.

Rad detaljnije opisuje osnovne probleme i modele za merenje efi-

kasnosti u pojedinim oblastima. Testovi i rezultati modela pokazuju izu-

zetan značaj izbora ulaznih i izlaznih promenljivih.

Ključne reči: efikasnost, logistika, metode merenja, Data Envelopment

Analysis.

Datum prijema članka/Paper received on: 31. 03. 2012.

Datum dostavljanja ispravki rukopisa/Manuscript corrections submitted on:

23. 04. 2012.

Datum konačnog prihvatanja članka za objavljivanje/ Paper accepted for publis-

hing on: 25. 04. 2012.