RFId and Returnable Transport Items Management: An

Activity-based Model to Assess the Costs and Benefits in

the Fruit and Vegetable Supply Chain

Giovanni Miragliotta, Alessandro Perego and Angela Tumino

Department of Management, Economics and Industrial Engineering

Politecnico di Milano, 20133 Milano, Italy

Abstract. Radio Frequency Identification (RFId) technology promises to enable

substantial benefits in Returnable Transport Items management. Lured by this

opportunity, some companies (e.g. Metro Group) are carrying out pilot projects,

but the difficulties in quantifying the costs and benefits stemming from such

applications still prevent many companies from using this technology. This

paper describes an analytical model to assess the profitability of such

investments, focusing on the fruit and vegetable supply chain.

1 Introduction

Radio Frequency Identification (RFId) technology is deemed to have great potential

to improve the efficiency and accuracy of asset management in different supply

chains. More specifically, the management of Returnable Transport Items (RTIs) - i.e.

“all means to assemble goods for transportation, storage, handling and product

protection in the supply chain which are returned for further usage, including for

example pallets with and without cash deposits as well as all forms of reusable crates,

trays, boxes, roll pallets, barrels, trolleys, pallet collars and lids” [1] - has always been

a critical issue in the Fast Moving Consumer Goods supply chain. Therefore, a

substantial interest in RFId potentialities has arisen, and some pilot projects have been

launched by supply chain leaders, e.g. Metro Group [2]. However, many companies

are still biding time, primarily because of a lack of confidence in the benefits and,

consequently, in the impact on Return On Investment, which generates a greater

perception of risk [3].

Consistent with this premise, this paper aims to present an activity-based model to

evaluate the profitability of RFId applications supporting RTI management in the fruit

and vegetable supply chain. The paper is structured as follows. Section 2 provides a

classification of the main scientific contributions on the evaluation of RFId

applications. Section 3 presents the objectives and the methodology. and Section 4

describes the model. Section 5 illustrates the application of the model to the supply

chain of a prominent Italian retailer and discusses the main business implications.

Finally, Section 6 draws some conclusions and suggests future research paths.

Miragliotta G., Perego A. and Tumino A. (2009).

RFId and Returnable Transport Items Management: An Activity-based Model to Assess the Costs and Beneﬁts in the Fruit and Vegetable Supply Chain.

In Proceedings of the 3rd International Workshop on RFID Technology - Concepts, Applications, Challenges , pages 39-48

DOI: 10.5220/0002192800390048

 SciTePress

2 Literature Review

The assessment of the value related to the implementation of ICT solutions has

fascinated for years researchers, and various studies have been conducted. However,

most of them have been carried out at the industry and economy levels, or at the firm

level, providing limited support to understand where and how the benefits are

generated. Only few authors analyse the impact of ICT on business processes, which

is essential to understand how value is created (e.g. [4]). If we focus on RFId, in

recent years members of both academic and generalist press have devoted much

attention to the evaluation of RFId projects. The available contributions can be

classified in three main groups.

1. Qualitative papers that describe the strategic implications of applying RFId to

supply chain management (e.g. [5]), provide a taxonomy and a qualitative

evaluation of the benefits achievable through RFId technology (e.g. [6], [7]) and

analyze the implementation process (e.g. [8]).

2. Quantitative analyses based on empirical evidence (e.g. case studies) or experts’

evaluations, which mainly focus on the Fast Moving Consumer Goods supply

chain, where the first pilot projects were launched. These papers seek to provide

both a taxonomy and a quantitative evaluation of the benefits stemming from the

adoption of RFId (e.g. [9], [10]).

3. Quantitative studies based on structured assessment models, that present

mathematical and simulation models to assess the impact of RFId on supply chain

performance (e.g. [11], [12], [13]).

While the first attempts were mainly based on empirical evidence, in recent years

more attention has been paid to mathematical and simulation models. However, they

mainly focus on the benefits that can be achieved applying the RFId tags on the

products (i.e. pallet loads, cases, items), while little attention has been paid to the

benefits enabled by the tagging of RTIs (e.g. [9]).

3 Objectives and Methodology

Consistent with the premises, this paper aims to present a structured model to assess

the costs and benefits stemming from the RFId tagging of RTIs, and to eventually

inspire the decision making process of managers. More specifically, the model

focuses on the management of the RTIs which are commonly used to transport fruit

and vegetables in the Fast Moving Consumer Goods supply chain. The structure of

the model is general, i.e. it can be applied to any retailer who uses RTIs to transport

fruit and vegetables, and consists of three strictly related components: the assessment

of the benefits in productivity, the assessment of the benefits enabled by the increased

process quality and visibility, and the assessment of the capital and operational

expenditures (CapEx and OpEx). The paper present the results of the model

application to a prominent Italian retailer (Nordiconad).

The research program was divided into three phases, which resorted to ad hoc

methodologies. In the first phase, direct interviews with the involved companies

(producer, retailer, pooling company) were performed in order to analyze the

processes. In the second phase, the analytical model was developed, resorting to the

well-established activity-based modeling approach, as used by other authors (e.g.

[14]). More specifically, in order to understand the impact on business processes, each

activity was first split into a hierarchy of elementary activities. Then, the impact of

RFId in terms of reduced resource requirements based on relevant operational

resource consumption drivers was evaluated. In order to improve its usability, the

model has been coded into an MS-Excel based tool. Finally, direct interviews and

technological tests were carried out to collect the required inputs and analyze the

results.

4 The Model

4.1 The Reference Supply Chain

The typical fruit and vegetable supply chain consists of four members: the pooling

company, the producer, the retailer (distribution center, stores) and the carrier. (cf.

Figure 1).

Pooling company

Producer

Retailer point of saleRetailer point of sale

Carrier

Retailer DC

Focus of the assessment model

Fig. 1. The reference supply chain.

The model considers two kinds of RTI, one being used as secondary packaging

(plastic crates and bins), while the other as tertiary packaging (pallets, roll containers).

Other relevant characteristics of this flow are as follows:

• Crates and bins are provided by a pooling company which sends them to the

producer, so that the latter can fill them with fruit and vegetables. In accordance

with the retailer’s orders, the producer sends the RTIs full of fruit and vegetables

to the retailer DC, who temporarily stores them. Then, the stores order the

products, and the retailer DC picks the requested crates and bins and ships them to

the points of sale. Finally, the retailer DC is responsible for the collection of the

RTIs, and for their shipment to the pooling company (thus closing the loop);

• Both roll containers and pallets are used to ship the products from the retailer DC

to the stores, while the producer uses only pallets;

• The owner of the crates and bins is the pooling company, while the pallet and roll

containers are owned by the retailer;

• The carrier is responsible for the transport of RTIs from the retailer DC to the

stores.

More specifically, the model is focused on a sub-part of the previously illustrated

supply chain (cf. Figure 1) which consists of the retailer DC, the carrier and the

stores. For each of these nodes the model focuses on the RTI receiving, shipping and

administrative activities. The latter have been analyzed in depth and the associated

consumption of resources has been modeled using an activity-based approach (cf.

Section 3). The results have been extensively reviewed and validated by the logistics

directors of the involved companies.

4.2 The Technological Scenarios

We considered an RFId scenario in which every RTI is provided with a re-usable

RFId tag. The latter is initialized when the asset enters into the system according to

the Electronic Product Code (EPC) standard, which states that a unique identifier has

to be written on the tag. The costs of the RFId tags are sustained by the company who

owns the RTIs, i.e. the pooling company for the crates and bins, and the retailer DC

for the pallets and roll containers.

According to the current performance of RFId technology, a non-zero reading

time has been assumed to ensure perfect reliability, i.e. 100% reading rate (cf. Section

5.1).

All the supply chain nodes are provided with the RFId hardware and software

required to identify all the entering and exiting RTIs. More specifically:

• The loading and unloading docks of the retailer DC are provided with RFId gates

consisting of an RFId reader, four antennas, a movement sensor, a traffic light to

confirm the correct reading, and a monitor;

• The stores are provided with RFId handheld readers;

• All the nodes are provided with appropriate software which collects the data

gathered by the hardware infrastructure.

As a benchmark, we chose a base-line scenario (B) which assumes a visual, non-

automatic identification. In fact, while the bar-code standard is commonly used to

identify the goods, it is still less used for RTI identification and management.

Moreover, the number and typology of shipped RTIs is written on the shipping

documents manually or electronically. As for the plastic bins and crates, the retailer

has to collect information on their usage, which is sent to the pooling company.

4.3 The Assessment of the Benefits

As stated in Section 3, two main classes of benefit were considered. Indeed, we

included both benefits related to productivity increases within the material handling

and the administrative processes (called “operational benefits”), and benefits that

accrue through better process accuracy and visibility.

An activity-based approach was used to assess the operational benefits. In

particular, receiving and shipping activities, along with the related administrative

processes, have been split into a hierarchy of elementary activities. Then, the impact

of RFId in terms of reduced resource requirements has been assessed. For the sake of

illustration, the model regarding the “Shipping of full RTIs from the retailer DC to the

stores” will be detailed here (cf. Figure 2). After completing the picking activity, the

worker registers the number of RTIs that have been used. Then, two documents are

prepared and printed: the shipping document and a control document, which contains

the typology and the total number of RTIs that are going to be shipped to the stores.

Before the goods are loaded on the truck, a worker counts the number of RTIs that

have been used, and compares it with the data reported in the control document. If

needed, he corrects the document with the right number of RTIs. Thanks to RFId

technology, the assets counting can be automated. For the interested reader, the

complete model (i.e. description of the activities, inputs, formulas) is available on

request.

Macro-activity Activity Elementary activity Input Formula

Time to re

ister the data in the IS

(

t_IS

)

Workforce usa

e factor

(

)

Number of orders per

ear

(

n_ord

)

Time to find the shippin

document in the IS

(

t_find_sd

)

Workforce usa

e factor

(

)

Number of trucks exitin

from the retailer DC per

ear

(

n_truck

)

Time to print a document

(

t_pr

)

Workforce usa

e factor

(

)

Number of trucks exitin

from the retailer DC per

ear

(

n_truck

)

Time to print a document

(

t_pr

)

Workforce usa

e factor

(

)

Number of trucks exitin

from the retailer DC per

ear

(

n_truck

)

Time to move to the loadin

dock

(

t_ld

)

Workforce usa

e factor

(

)

Number of trucks exitin

from the retailer DC per

ear

(

n_truck

)

Time to count an RTI

(

t_RTI

)

Workforce usa

e factor

(

)

Number of RTIs shipped to the stores

(

n_RTI

)

RETAILER DC

Write the number of

RTIs in the Information

System (t_write_IS)

Registration in the

Information System

(t_write_IS)

t_ship_full = t_write_IS + t_doc + t_cont

t_write_IS = (t_IS * n_ord) / η

t_request = (t_find_sd * n_truck) / η

t_print_sd = (t_pr * n_truck) / η

t_print_cd = (t_pr * n_truck) / η

t_move = (t_ld * n_truck) / η

t_count = (t_RTI * n_RTI) / η

Move to the loading

dock (t_move)

Counting (t_count)

Controls (t_cont)

Shipping - Full

RTIs (t_ship_full)

Request to the

Information System

(t_request)

Print the shipping

document (t_print_sd)

Print the control

document (t_print_cd)

Preparation of the

shipping documents

(t_doc)

Fig. 2. The benefits evaluation – Shipping of full RTIs from the retailer DC to the stores.

As regards the benefits enabled by the increased process accuracy and visibility, a

brief description of those which have been quantified is reported here.

• Shrinkage. RFId-enabled automatic identification helps reduce process errors and

improve asset visibility, leading to a reduction in penalty costs when an RTI is not

found. In fact, when an RTI is lost, the supply chain node which is in charge of it

has to pay a pre-defined amount of money to the asset owner.

• Contentious Issues. “Lost RTIs” means that contentious issues are likely to arise

between the involved nodes. In fact, it is not easy to understand when and where

the loss occurs, since currently the RTIs are not individually tracked. The RFId-

enabled automatic identification of all the RTIs supports their tracking and tracing,

thus reducing the insurgency of contentious issues.

• Better Process Control. The automatic identification allows to introduce new

counting controls, which today are not performed because of time (and cost). This

benefit has been assessed in terms of amount of time (and related costs) which

would have been needed in the base-line technological scenario to perform the RTI

counting enabled by RFId.

In addition to the benefits that can be quantitatively assessed, others are of a more

qualitative nature. For example, the RFID-enabled tracking and tracing of all the RTIs

allows to create a database which contains the current location of all the assets. This

allows to improve the rotation of RTIs, as well as to better plan their maintenance.

4.4 The Assessment of Capital and Operational Expenditures

The implementation costs of an RFId project include both the initial investment

(Capital Expenditure, CapEx) and the recurrent annual costs (Operational

Expenditure, OpEx). More specifically, CapEx includes the costs of hardware (e.g.

readers, antennas, re-usable tags), software (middleware and software

development/integration), and project management (design, implementation, test and

change management, project management). The OpEx includes the maintenance of

the RFId infrastructure and the annual training costs.

5 The Application of the Model

5.1 The Parameters

The model has been applied to the fruit and vegetable supply chain of a prominent

Italian retailer, i.e. Nordiconad. The main flows which characterize the supply chain

are reported in Table 1.

Table 1. The RTI flows (thousands of RTIs).

Retailer DC Carrier Store

C1 C2 C3 C1 C2 C3 C1 C2 C3

Inbound pallets

(kpallets/year)

Full

30 30 60 35 115 210 35 115 210

Empty

35 115 210 35 115 210 - - -

Outbound

pallets

(kpallets/year)

Full

35 115 210 35 115 210 - - -

Empty

30 30 60 35 115 210 35 115 210

Inbound rolls

(krolls/year)

Full

- - - 145 460 860 145 460 860

Empty

145 460 860 145 460 860 - - -

Outbound rolls

(krolls/year)

Full

145 460 860 145 460 860 - - -

Empty

- - - 145 460 860 145 460 860

Inbound bins

(kbins/year)

Full

6 6 11 6 6 11 6 6 11

Empty

6 6 11 6 6 11 - - -

Outbound bins

(kbins/year)

Full

6 6 11 6 6 11 - - -

Empty

6 6 11 6 6 11 6 6 11

Inbound crates

(kcrates/year)

Full

4,20

8,00

4,20

8,00

4,20

8,00

Empty

4,20

8,00

4,20

8,00

- - -

Outbound

crates

(kcrates/year)

Full

4,20

8,00

4,20

8,00

- - -

Empty

4,20

8,00

4,20

8,00

4,20

8,00

Three different scopes of analysis have been considered, which differ in terms of

costs and benefits:

• Case 1 (C1), in which the RFId project involves only the sub-part of one retailer

DC that is in charge of managing the fruit and vegetables, and the carriers and

stores who receive these products. The RTIs which are used to transport all the

other products are not provided with RFId tags, and the related activities are not

considered.

• Case 2 (C2), in which the RFId project involves one retailer DC, which adopts the

technology to support not only the management of the RTIs used to transport fruit

and vegetables, but also of those which are used to transport all the other grocery

products. All the carriers and stores who receive products from the considered DC

are included in the analysis.

• Case 3 (C3), in which the retailer adopts RFId technology in all its DCs to support

the management of all RTIs (both those which are used to transport fruit and

vegetables and those which are used for other grocery products), including the

DCs which do not manage fruit and vegetables. All the carriers and stores who

receive the products from the retailer DCs are considered.

The profitability analysis compares the RFId scenario with a base-line scenario

(B), whose main performance parameters are reported in Table 2. As regards the RFId

scenario, two different identification solutions can be used accordingly to the facility

needs, namely RFId gates or handhelds. For sake of illustration, the RFId gates are

used to identify the RTIs entering in (or exiting from) the retailer DCs, while the RFId

handhelds are preferred in the points of sale because of their higher flexibility. These

two solutions differ in terms of time to identify the RTIs (cf. Table 2).

Table 2. The main performance parameters.

Base-line

RFId

Gate

Handhel

Time to identify all the crates/bins on a pallet/roll (s/pallet;

s/roll)

145 7.5 15

Time to archive the control documents (minutes/day) 90 0

% errors in the control documents prepared by the retailer DC 2% 0%

% errors in the control documents prepared by the store 5% 0%

5.2 The Results

Figure 3 shows the benefits enabled by the RFId adoption, for the three considered

scopes of analysis (“Case 1”, “Case 2”, “Case 3”).

First, the RFId adoption allows to achieve both significant productivity benefits,

which range from 260,000 €/year in “Case 1” to 880,000 €/year when all the products

and distribution centers are impacted by RFId (“Case 3”). Second, it yields also

substantial benefits enabled by the increased process accuracy and visibility, which

range from about 100,000 €/year in “Case 1” to 475,000 in “Case 3”. Therefore, the

total benefits range from 360,000 €/year to more than 1,350,000 €/year.

Interesting results emerge from the breakdown of the benefits by activity and by

supply chain stage. Overall, the supply chain members enjoy different benefits. More

specifically, it is the retailer DC that reaps most of the benefits, significantly reducing

the costs of both receiving the empty RTIs from the stores and performing

administrative activities. Shrinkage and contentious issues reduction is significant too.

As regards the stores, in the base-line scenario the most critical activity is to ship the

empty RTIs to the retailer DC, since every error would lead to a contentious issue.

Therefore, most of the benefits are related to the automatic counting and the

contentious issues reduction. The benefits achieved by the carrier, instead, are mainly

related to the possibility of eliminating the controls performed when receiving the

RTIs, because RFId technology enables a more objective and accurate measure of the

number of received assets.

Looking at the supply chain as a whole, the receiving and administrative activities

reap most of the total benefits. The contentious issues reduction and the improved

process control are significant too.

Benefits: Breakdown per SC player

200

400

600

1400

Case 1 Case 2

Case 3

Supply Chain

k€/year

358,5

Benefits: Breakdown per activity

Receiving

Shipping

Administrative

activities

Retailer DC

Stores

Carrier

800

1000

1200

208

102,5

681

398

147

136

1354

745

295

314

200

400

600

800

1400

Case 1 Case 2

Case 3

Supply Chain

k€/year

358,5

132

1000

1200

Shrinkage

Contentious issues

Process control

28,5

681

210

203

103

1354

380

403

206

188

SC players

Activities

Fig. 3. The results of the model – The benefits in the RFId scenario.

Table 3 reports the Capital and Operational Expenditures incurred by the supply

chain members.

Table 3. CapEx and OpEx in the RFId scenario.

Retailer DC Stores Supply Chain

C1 C2 C3 C1 C2 C3 C1 C2 C3

CapEx (k€) 733 827 1,738 214 275 584 947 1,102 2,322

RFId HW – Tags 685 690 1,350 - - - 685 690 1,350

RFId HW – Other 37 114 328 190 230 486 227 344 672

RFId SW 10 20 50 15 35 75 25 55 105

Training 1 3 10 9 10 23 10 13 23

OpEx (k€/year) 45,6 82,3 165,5 18 25 52 63,6 107,3 217,5

Maintenance – HW 30 65 130 15 18 39 45 83 169

Maintenance – SW 1,5 3 7,5 2,2 6 11 3,7 9 18,5

RFId tags on new RTIs 14 14 27 - - - 14 14 27

Training to new people 0,1 0,3 1 0,8 1 2 0,9 1,3 3

As stated before, each RTI is provided with a re-usable RFId tag, whose cost

depends on the specific RTI. More specifically, the tags which are put on pallets,

crates and bins cost about 0.2€/tag, while special tags are used on the “metallic” roll

containers (10 €/tag).

The payback time and the Net Present Value (NPV) have been computed, as

shown in Table 4, taking into account a “transitory period” (1 year) during which we

have assumed that the benefits can only be partially achieved. The Net Present Value

has been computed using a five-year time horizon and a discount rate of 4%, whereas

the payback time is based on non-discounted cash flows.

If we consider only those RTIs used to transport the fruit and vegetable flows of a

single distribution center (“Case 1”), a positive Net Present Value cannot be realized

from a supply chain perspective, because only the carrier benefits from the technology

adoption. But when we consider all the RTIs used by one (“Case 2”) or more (“Case

3”) retailer DCs to transport all the grocery products, the investment becomes

profitable from a supply chain perspective. Moreover, Table 4 shows that the best

adoption strategy for the retailer is to adopt RFId in one Distribution Center, which

manages both fruit and vegetables and other products. The results obtained by the

retailer DC in “case 3” are only apparently inconsistent. We could expect the

investment profitability to be similar to that calculated in “Case 2”, but the lower

investment profitability in “Case 3” is due to the fact that some of the additional DCs

included in the analysis do not manage fruit and vegetables (cf. Section 5.1), which

require most of the efforts in the base-line scenario.

Table 4. The investment evaluation – NPV and Pay-back time.

Retailer DC Carrier Stores Supply Chain

C1 C2 C3 C1 C2 C3 C1 C2 C3 C1 C2 C3

NPV (k€)

-263 180 -33 138 385 755 -27 -105 -112 -151 460 610

Pay-back

time (years)

7.7 3.7 6.7 0 0 0 5.3 8.0 5.8 5.6 3.1 3.6

A sensitivity analysis has been carried out in order to analyze how the results

change in function of the most significant parameters. In particular, different reading

time performance of the RFId technology (± 100%) and different costs of the RFId

tags (± 20%) have been considered. The analysis proved that, considering a lower

performance level and a higher tag costs, the investment remains profitable in “Case

2” and “Case 3”, with a pay-back time of 4.0 and 4.7 years respectively. Moreover,

when best technology performance and lower costs are considered (which can jointly

be obtained thanks to a technological improvement and/or a more accurate analysis of

the best tagging position), the investment becomes profitable also in “Case 1”, with a

pay-back time of about four years.

6 Conclusions

This paper shows that RFId technology has the potential to improve the efficiency and

accuracy of the Returnable Transport Items in the fruit and vegetable supply chain.

The results are strongly dependent on the adoption level in terms of facilities and

products which are supported by RFId technology (“Case 1” vs. “Case 2” vs. Case

3”). In fact, while the application of RFId to support only the management of RTIs

which are used to transport fruit and vegetables seems to be unprofitable, a positive

Net Present value is obtained when an entire facility is considered.

The model presented in this paper has a few limitations. The most important is

that it does not consider the pooling company and the producer, both of whom could

significantly benefit from RFId adoption. Moreover, a wider sensitivity analysis

should be performed, since only the identification time has been considered so far.

Moreover, the application of the model to other retailers would allow to compare the

project profitability in different real contexts, thus providing interesting information.

Finally, only some of the accuracy and visibility benefits have been quantitatively

assessed. Future developments of the research should overcome these limitations.

Acknowledgements

We would like to thank GS1 Italy and Nordiconad for their valuable support during

all the phases of the research project.

References

1. ISO: ISO TC104 – Supply chain applications of RFId – Returnable Transport Item,

ISO/CD 17364, ISO 2005

2. Wolfram G.: RFId/EPC value-added applications in the supply chain at Metro Group,

RFId/EPCglobal in Europe, Düsseldorf, 2006

3. RFId-IPO, Italian Permanent Observatory on RFId Technologies: RFId: alla ricerca del

valore, research report (in Italian), 2007

4. Mooney J.G., Gurbaxani V., Kraemer K.L.: A Process Oriented Framework for Assessing

the Business Value of Information Technology, Proceedings of the 16th Annual

International Conference on Information Systems, 2001

5. Srivastava B.: Radio Frequency ID technology: the next revolution in SCM, Business

Horizons, Vol. 47, No. 6 (2004), 60-68

6. Helders B., Vethman A. J.: Beyond 2005: how RFId will change the global Supply Chain,

Chain Store Age, Vol. 79, No. 12 (2003), 39-48

7. Curtin J., Kauffman R.J., Riggins F.J.: Making the ‘MOST’ out of RFId technology: a

research agenda for the study of the adoption, usage and impact of RFId, Information

Technology and Management, Vol. 8 (2007), 87-110

8. Wu N.C., Nystrom M.A., Lin T.R., Yu HC.: Challenges to global RFId adoption,

Technovation, Vol. 26, No. 12 (2006), 1317-1323

9. Chappell G., Ginsburg L., Schmidt P., Smith J., Tobolski J.: Auto-ID on the Line: The

Value of Auto-ID Technology in Manufacturing, Auto-ID Center, MIT, 2003

10. Hardgrave B.C., Waller M., Miller R.: Does RFId reduce out of stocks? A preliminary

analysis”, ITRI, University of Arkansas, 2006

11. Atali A., Lee H., Ozer O.: If the inventory manager knew: value of RFId under imperfect

inventory information, Stanford University, 2004

12. Wang S., Liu S., Wang W.: The simulated impact of RFId-enabled supply chain on pull-

based inventory replenishment in TFT-LCD industry, International Journal of Production

Economics, Vol. 112, No. 2 (2008), 570-586

13. De Kok A.G., van Donselaar K.H., van Woensel T.: A break-even analysis of RFId

technology for inventory sensitive to shrinkage, International Journal of Production

Economics, Vol. 112, No. 2 (2008), 521–531

14. Subirana B., Eckes C., Herman G., Sarma S., Barrett M.: Measuring the impact of

information technology on value and productivity using a process-based approach: the case

of RFId technology, MIT, 2003