Improving the collection process of the returnable transport packaging at Phoenix Beverages Limited

(1)

October 2011 Master Thesis – Technology Management

Improving the

collection process of

the returnable

transport packaging at

Phoenix Beverages

Limited

The differential deposit refund system

(2)

2

Improving the collection process of the returnable

transport packaging at Phoenix Beverages Limited

‘The differential deposit refund system’

Education:

MSc. Technology Management

Faculty of Economics and Business

University of Groningen

Supervisor RUG:

H. Van der Meulen, MSc.

Co-assessor RUG: dr. N.D. Van Foreest

Supervisor PBL:

H. Rivet, supply chain manager

Author:

René van der Meer

Student number:

1730975

(3)

3

Acknowledgement

This master thesis is written for finalizing my study of MSc. Technology Management at the University of Groningen. I had the great opportunity to conduct this research at Phoenix Beverages Limited (PBL) on the tropical island of Mauritius, in the Indian Ocean. This research focused on the collection process of the returnable transport packaging. During this period I met some great people who helped me to finish this research successfully. For that reason, I want to pay a tribute to all who helped me one way or the other during this period.

First, I would like to thank Hugues Rivet for offering me the opportunity to perform my research at PBL. I would also like to thank my other colleagues and friends Sarind Bundhoo and Yogen Vencatapillay for having a great time at PBL and support they gave me to succeed this research.

From the university I wish to acknowledge my supervisor MSc. Hendrik van der Meulen for his support, guiding, feedback and our cooperation during this period. Due to his supervision I kept the academic level in mind. Secondly I would like to thank dr N. D. van Foreest for providing his co-assessor work.

To my parents, I owe everything since none of this would happen if they not supported me during my study.

(4)

4

Management summary

Phoenix Beverages Limited is a beverage company that engages in alcoholic and non-alcoholic beverages which is located on the island Mauritius. It sells a substantial amount of it beverages in returnable transport packaging (RTP), including returnable glass bottles (RGB), as well as kegs and crates. However the last couple of years the company seems to have a huge problem of getting all these RTP back to the company. This results in short supply of RTP, loss of sales and money. The objective of this thesis is to perform a research for finding the problem(s) and their root cause(s) and improve their collection process of the RTP. In order to achieve this objective, a research questions is established:

“How should PBL deal with the root causes in order to improve their collection process of the returnable transport packaging towards a capture rate of 98%?”

The answer is given by finding and identifying the problems, discussing possible potential projects, and analyzing a new collection process design.

The base of this research is structured in three phases according to the DOV-model (i.e. Diagnostic, Design, Change) of the Leeuw (2003). Furthermore, these phases are supported with the use of key activities in understanding and improving existing processes and key activities in designing and prototyping a new process given by Davenport (1993).

(5)

5 To deal with these problems a new process design for the collection process is analyzed, the differential deposit refund system (DDRS). The DDRS is an operation system that makes use of a deposit and a handling fee. To ensure the return of RTP, PBL makes use of a deposit. This is an inducement to compensate the consumer of returning the RTP. But, the deposit does not compensate the retailer for the inconvenience (i.e. handling, time and loss of space) that is caused by accepting the RTP. Hence, the intention of using a handling fee is to compensate the retailer for the handling and loss of space and time.

By doing this the problem of RTP acceptance and the causes time and handling could be dealt with. The retailer could be encouraged to accept more RTP from the consumer. Furthermore, the customer could be encouraged to do better sorting. This result in more RTP acceptance by the salesmen and more RTP will be captured.

With the use of „what-if‟ analysis the DDRS project is „prototyped‟ to simulate and test the operation in order to learn and reduce implementation risks for possible future developments. Based on future developments different scenarios are made which are build around two critical factors;

- Value of the handling fee and;

- The increase of the acceptance at the retailer. These factors determine the value of the assets;

- Cash flow and; - Product flow.

The handling fee is funded with the unclaimed deposits (i.e. deposit price – purchase costs). In all of the scenarios a positive cash flow is gained vary from Rs 398.996 to Rs 13.312.101. From these scenarios, it can be concluded that the unclaimed deposits can be used to fund the full handling fee. Because of the increase in product flow, cost savings can be realized (i.e. investment in RTP, missed sales and intern sorting costs) as from Rs 17.066.375 to Rs 81.427.678 compared to the current situation.

(6)

6

List of abbreviations

ABC-method : Activity-Based Costing

Bts : Bottles

CM : Conceptual Model

CU : Commercial Unit

DDRS : Differential Deposit Refund System

DOV-model : Diagnosticeren, Ontwerpen and Veranderen (or Diagnostic, Design, and Change) ECM : Empirical Causal Model

MBL : Mauritius Breweries Ltd. MDC : Managed Distribution System PBL : Phoenix Beverages Limited PCM : Phoenix Camp Minerals PET : Polyethylene Terephthalate RGB : Returnable Glass Bottles

Rs : Rupees

(7)

7

1. Introduction

1.1 The company: Phoenix Beverages Limited

Phoenix Beverages Limited (PBL) is a beverage company that engages in the brewing of beer, as well as in bottling and of selling beer, wines, spirits, sparkling drinks, soft drinks, table water, and alternative beverages in Mauritius and internationally.

PBL was born from the strategic merger of Phoenix Camp Minerals (PCM) and Mauritius Breweries Ltd (MBL) in 2001 to reinforce its position on the local and regional market. PCM was established in 1931 and was the sole bottler of Coca-Cola Company Ltd in Mauritius. MBL was established in 1961 and its flagship brand, the Phoenix Beer. Today, PBL ranks amongst the top companies of Mauritius, where it is the leading beverages company.

PBL also produces and sells beverages under contract agreement with Diageo, which include brands such as Guinness Foreign Extra Stout, Malta Guinness, and Smirnoff Ice.

The company employs over 1100 employees working on four different facilities; - the brewery;

- the soft drink plant commonly known as the „Limonaderie‟ (or Limo); - the Commercial Unit (CU) and;

- the Head Office.

The brewery and Limo produces an average of 1.4 million hectoliters of alcoholic and non alcoholic drinks and delivers its products over 9000 outlets and have a turnover of Rs 3 billion.

The brewery has a brewing capacity of around 550,000 hectolitres per annum and three packaging lines; two glass returnable lines and one canning line. The majority of production is for the local market. But PBL is developing rapidly its export market, mainly in Reunion Island where its market share has grown from less than 2% to more than 10% during the last three years. PBL also exports beer to Australia, UK, Indonesia and some countries in Africa.

The soft drink plant produces all Coca-Cola carbonated soft drinks and table water, and has five packaging lines; one for returnable glass bottles, two PET Lines, one bottled water line and another one for water jars. The total production capacity of the soft drink production lines is around 850,000 hectolitres per annum. PBL also does some contract manufacturing under license from Coca-Cola for some countries in the region, for instance it produces Coca-Coca-Cola in glass for Reunion Island and bottled water for Mayotte.

(9)

9 The warehousing operations at PBL includes the storage and movement of finished goods as from the end of the production lines until the goods leave the company‟s distribution centre for delivery to customers. The warehousing can be divided into three sub-processes which are:

- Forwarding; the transfer of goods from production sites to the Commercial Unit; - Storage and;

- Load Preparation for distribution.

These process operations play a crucial role in PBL‟s supply chain as logistics and distribution are a major competitive advantage for the company. PBL benefits from considerable economies of scale in the commercialization of their products and their portfolio allows the company to provide an attractive commercial offer. All orders (i.e. around 1200 daily) are taken one day prior to delivery, are processed, prepared (e.g. picked) and loaded onto trucks on the same day.

(10)

10

1.2 Problem definition

1.2.1 Background of the research

Phoenix Beverages Limited (PBL) sells a substantial amount of its beverages in returnable transport packaging (RTP), including returnable glass bottles (RGB) for alcoholic and non-alcoholic drinks, as well as kegs and crates. In 2010 approximately 40% of the sales volume was sold in RGB. The use of RTP absorbs substantial resources. Investments in crates, bottles, and kegs are needed to facilitate product introductions or to replace damaged and lost RTP. Also, the handling of RTP involves operations, such as transportation, storage, sorting, and the cleaning of used RGB and kegs, which are complementary to the brewery activities.

PBL is highly dependable on the RTP, especially on the RGB. The supply of glass is worldwide being under strain and due to its relative small demand in glass1; it is a real challenge for PBL to obtain the needed glass bottles to run its business (Rivet, 2009). Furthermore, the price of glass is increasing.

This collection process of the RTP from the customers (i.e. retailers and Managed Distribution Center), sorting and return to the production facilities is an essential and critical operation at the PBL. However, the last couple of years the company seems to have a huge problem of getting all these RTP back to the company.

This results in a short supply of RTP, and stock-outs at the brewery and soft drink plant causing disruptions of the bottling process for the production line which hampers the production scheduling. This results in missed sales and loss of money. According to Groenewold and van der Kley (2009) the total value of missed sales in 2007 was Rs 24 million and Rs 20 million in 2008 on annually basis. They did a research in the management of RTP at PBL.

To prevent short supply of RTP and stock-outs, PBL purchases new RTP. Figure 1 shows the RTP (i.e. RGB) purchases and sales as from 2006 until 2010. As can be seen from 2008, the investment in purchase costs and - quantity of new RGB is growing. However, the sales remain the same. The less RGB gets back to the company, the more new RTP needs to be purchased to meet up the demand.

(11)

11 FIGURE 1.RGB PURCHASES AND SALES AS FROM 2006 UNTIL 2010

1.2.2 Problem statement

The company seems to have a huge problem of getting all the RTP back to the company. They do not know what causes this collection process problem. Therefore, in PBL‟s there is a need to go into the details and to figure out the problems and their root cause(s) of this collection process problem. The outcomes can be used to improve the collection process of the RTP. Based on the background of the research the objective of this study is:

“To perform a research on the collection process within the Phoenix Beverage Limited company for finding the problems and their root cause(s) and improve the collection process of the returnable transport packaging”.

This research should help to run PBL its business by preventing the short supply of RTP and stock-outs at the brewery and soft drink plant. Preventing stock-stock-outs results in less disruption at the bottling process for the production line. The production scheduling will not be hampered and fewer sales will be missed. Thereby less investment has to be done to replace the lost RTP.

To evaluate the process objective of the collection process a measure of performance should be used. According to Beullens et al. (2002) the first step in handling the return of RTP is the effective acquisition from the customer. By periodically measuring the capture rate, i.e. the information about

0 1000000 2000000 3000000 4000000 5000000 6000000 7000000 Rs. -Rs. 200.000 Rs. 400.000 Rs. 600.000 Rs. 800.000 Rs. 1.000.000 Rs. 1.200.000 Rs. 1.400.000 Rs. 1.600.000 2006 2007 2008 2009 2010

RGB purchases and sales

(12)

12 all incoming RTP and outgoing RTP (Van Dalen et. al, 2005), gives an idea of the effectiveness of the collection process. Based on this information, the capture rate can be measured in percents by the following formula [1].

[1]

The capture rate of April 2010 until March 2011 at PBL is illustrated in figure 2 which is derived from RTP tracking data2. The figure shows a lot of variation in capturing that took place. The capture rate varies from between 0% and 160%. A capture rate nearby the 0% occurs when only outgoing RTP takes place. A capture rate higher than 100% occur when there are more incoming RTP than outgoing RTP. On average a capture rate of 93% is obtained. However the standard deviation is 22% which shows that the capture rate varies during the year. This means, that the capture rate fluctuates a lot which makes the capturing of RTP unpredictable. This results in a short supply of RTP when the demand is high.

In total the incoming RTP is 4.880.248 cases and outgoing RTP of 5.234.725 cases per year. That means that 354.478 of cases are not coming back to the company. Investment needs to be done in order to facilitate the lost RTP. Appendix A shows the result of the purchasing costs and quantities of RTP types RGB and crates of year 2010. On average 395.122 units of RTP are purchased with a total cost of Rs 108.444.778. In ratio, replacing 354.478 units of RTP is a money loss of around Rs 97.020.431.

For the purpose of the objective an external benchmarking is done of different breweries with their capture rate, see appendix A. In the literature capture rates are found varying from 84 until 98%. Based on these capture rates and the overall objective of this study a research question is established:

“How should PBL deal with the root causes in order to improve their collection process of the returnable transport packaging towards a capture rate of 98%?”

To collect the needed information for answering the research question, the following sub-questions are established:

1. How can the problems be identified?

2. Which problem(s) and their root cause(s) have influence on the collection process? 3. How to deal with the root causes in order to improve the collection process? 4. How does the collection process look like in the new situation?

5. What is the effect of the new collection process at PBL?

2

(13)

13 FIGURE 2: THE PERFORMANCE OF APRIL 2010 UNTIL MARCH 2011 OF THE COLLECTION

PROCESS OF RTP -500000 -450000 -400000 -350000 -300000 -250000 -200000 -150000 -100000 -50000 0 0% 20% 40% 60% 80% 100% 120% 140% 160% 180%

Performance of the collection process

(14)

14

1.3 Research methodology

The previous section described the company, the background of the research and the problem statement. This section discusses the methodology applied for this research in order to answer the research question and sub-questions. The base of the research is structured into the three phases according to the DOV-model of de Leeuw (2003) (see figure 3 for a visual representation). DOV stands for the phases „Diagnosticeren, Ontwerpen, Veranderen‟, or Diagnostic, Design and Change. In the first phase, diagnostic phase, the causes of the problem situation will be clarified and used as a basis for the Design -and Change phase. The Design phase discusses the design alternatives (i.e. potential projects) that deal with the outcomes of the Diagnostic phase. In the Change phase the chosen potential project is simulated and analyzed.

Diagnostic Problem Design Solution Change

Problem situation Improved situation

FIGURE 3.DOV-MODEL OF DE LEEUW (2003)

1.4 Research overview

This research is organized as follows. In the first section of the research, the diagnostic phase, sub-questions 1 and 2 will be answered. A study of related literature is done in order to have better understanding and insight of this specific collection problem. The factors identified in the literature study are discussed in the conceptual analysis and verified in the empirical analysis. This results in an empirical causal model which clarifies the problems and their root causes.

The second section of the research, the design phase, sub-question 3 will be answered. Possible potential projects are presented that could deal with one or more problems and their root causes described in the previous section. The potential projects are evaluated, prioritized and selected for further research.

The third section of the research, the change phase, the sub-questions 4 and 5 will be answered. The potential project will be simulated and analyzed in order to reduce implementation risks by considering possible future developments.

(15)

15 Evaluation

Chapter 5

Chapter 6

Change phase Chapter 4

Design phase Chapter 3

Introduction Chapter 1

Diagnostic phase Chapter 2

Background of the research Literature study Conceptual analysis Potential projects Influencing factors Project choice Empirical Causal Model Current state Performance Project selection matrix Operation of the project What-if analysis

Scenarios Asstes and factors

Conclusion Discussion and recommendation Reviewing Problem statement Methodology (DOV-model)

(16)

16

2. Diagnostic phase

In the diagnostic phase the problem situation will be clarified. Within the diagnostic phase questions 1 and 2 will be answered. In order to identify the problems and their root causes (i.e. sub-question 1) a literature study will be performed in order to have better understanding and insight of this specific collection problem. Different factors can be found that could have influence on the performance of the collection process which results in a conceptual model.

The outcome of the diagnostic phase will be used as a basis for the Design –and Change phases. In order to understand an existing process the following key activities will be used in the diagnostic phase (Davenport, 1993):

- Describe the current process flow;

- Measure the process in terms of new process objectives; - Asses the process in terms of the new process attributes; - Identify problems with or shortcomings of the process; - Identify short-term improvements in the process;

- Asses‟ current information technology and organization.

The current process flow (i.e. collection process) will be described with respect to the conceptual model that is derived from the literature study. By doing this, the process can be measured in terms of new process objectives and assessed in terms of new process attributes. Process attributes might be considered as principles of a process operation. It addresses both high-level process characteristics and specific enablers (Davenport, 1993). Thereafter, the problems and their root causes can be identified. This results into an empirical causal model (ECM) which clarifies the problems and their root causes of the collection problem (i.e. sub-question 2). The ECM could help to identify short-term improvements in the collection process. The current organization will only be assessed if it is relevant for this research. Assessment of the organization includes job description, skills inventory, and knowledge of recent organizational changes (Davenport, 1993).

2.1 Literature study

For the purpose of this research, a discussion of related literature will be done. The objective of this literature study is to identify a set of factors that could clarify the problem situation.

(17)

17 forward in the chain, will be referred as the returnable transport packaging (RTP). RTP that goes back and forward in the supply chain constituting reverse flow in the logistics, better known as the term reverse logistics.

The term of reverse logistics is used and defined differently by different authors. The most common definition that is used by authors is by Rogers and Tibben-Lembke (1999), they define reverse logistics as: “The process of planning, implementing, and controlling the efficient, cost effective flow of raw materials, in-process inventory, finished goods and related information from the point of consumption to the point of origin for the purpose of recapturing value or proper disposal. In the paper by de Brito and Dekker (2003) different definitions of the term reverse logistics are compared and defined. According to them, the conception of reverse logistics dates from a long time ago, and the denomination of the term is difficult to trace with precision. De Brito and Dekker (2003) introduced the RevLog definition which describes the reverse logistics in a rather general term as: “The process of planning, implementing and controlling flows or raw materials, in process inventory, and finished goods, from the point of use back to a point of recovery or point of proper disposal”3

. In order to handle RTP in practice, several decisions about the logistics process of transport packaging have to be taken. According to de Brito (2003), the general process of handling the return of RTP exist out of three stages:

1. Collection;

2. Receipt & sorting in the warehouse and; 3. Storage in the warehouse;

Figure 5 gives an overview of the process of handling the return of RTP in its simplest form.

Return handling (1) Collection (2) Receipt & Sorting in the

warehouse (3) Storage in the warehouse Reuse

FIGURE 5. PROCESS OF HANDLING THE RETURN OF RTP(DERIVED FROM DE BRITO,2003&

DE KOSTER,2001)

The first stage a company has to decide whether too simultaneously pick-up RTP and distribute new RTP (de Brito, 2003). According to de Brito (2003), the trade-off here is between the exploitation of existing resources (i.e. trucks and route planning) and the increase of complexity (e.g. new and returned products sharing truck capacity). After stage 1, the RTP needs to be returned to the company and will be receipt and sorted in the warehouse. The company has to decide whether or not these have

(18)

18 to be done in the same area or facility where procured RTP are handled (de Brito, 2003). In stage 3 the RTP in good condition are put back in inventory to be sold or reused again (de Brito, 2003).

According to Beullens et al. (2002) major concerns in reverse logistics are both the collection and movement of the RTP. They did a study about the collection and vehicle routing issues in reverse logistics in which they discuss the importance of different available design options and functions a collection system should meet. They make a distinction between four aspects:

1. Collection infrastructure; 2. Collection policy;

3. Combination of level of the collection, and; 4. Characteristics of the collection vehicles.

The collection infrastructure relates to the points at which generators (e.g. customers) handover the used products (e.g. RTP) to the network. The collection policy specifies the moment at which a collection point is serviced and the volume per collected visit. The combination level is about services which are related to different classes of flows of goods that can be combined in various ways. The characteristics of the collection vehicles have to match the collection infrastructure, collection policy and combination level.

In order to manage the RTP different systems can be used. Kärkkäinen et al. (2004) focused on different approaches to managing the rotation of RTP based on earlier work by Lützebrauer, Kroon and Vrijens (1995). They distinguished three types of return logistic systems:

1. Switch pool systems;

2. Systems with return logistics, and; 3. Systems without return logistics.

They used nine company cases to describe and understanding different systems for managing the rotation of RTP. The return logistic systems were classified by the attributes (Kärkkäinen et al. 2004):

- Ownership; shows the owner(s) of the RTP in each of the rotation systems; - Return; is the party who takes care of delivering the RTP back to their storages;

- Storage and maintenance; present the responsibilities for warehousing, as well as repair and cleaning operations;

- Control; lists the participant(s) keeping count on the RTP;

- Deposit/rent; indicates whether a deposit is required from the recipients of the RTP, or they charged a rent;

(19)

19

2.2 Conceptual analysis

Based on the studies done by de Brito (2003), Beullens et al. (2002) and Kärkkäinen et al. (2004) a conceptual model is made (see figure 6). This conceptual model gives an overview of which factors could clarify the problem and shows (possible) causal relationships. The conceptual model helps to understand the knowledge that is needed to answer the research question. It should be noted that this model does not have to be regarded as a complete picture of the problem situation. Significant factors that arise during this research will be taken into account.

2.2.1 Description of the conceptual model

The research question address the collection process. This research will emphasis on the first stage of handling the return of RTP, which is the process stage of collection. Collection refers to all activities rendering products available and physically moving those to some point for further treatment (Fleischmann et al., 2000). According to Beullens et. al (2002) this involves aspects about the collection infrastructure, the collection of policy, the combination of level of the collection, and the characteristics of the collection vehicles. To manage RTP different return systems can be used with respect to the attributes given by Kärkkäinen et al. (2004).

The deposit is an attribute that could manage to unsure the return of RTP. The deposit act as an inducement to compensate the consumer for the inconvenience of returning the RTP (Fisher and Horton, 1979). According to Grimes-Casey et al. (2006) the deposit must be large enough to make return worthwhile to consumers, but not too much that it discourages sales. This means that the return of RTP could be influenced by the value of the deposit. This involves the aspect of the collection infrastructure. Without the return of the RTP, no handover to the network could take place.

Moreover the collection infrastructure involves a network whereby it can be very complicated and confusing for both retailers and manufacturers as they are dealing with multiple parties (i.e. PBL, MDC‟s, salesmen and other supply chain parties) (Schatteman, 2003). This means that there are usually many points from which RTP have to be collected. In other words stated by Fleischmann (2000), “the one-to-many structure that characterizes the forward distribution of goods becomes a many-to-one instead”. According to de Brito (2003) it will make a difference whether the end-user (i.e. customer) is an individual or an institution (e.g. bulk-use). This demands different locations for collection of different degrees of effort from the end-user (e.g. bring the product to a collection point).

(20)

20

Aspects of functions & designs Attributes of managing

Performance Collection infrastructure Combination level Vehicle type Collection policy Deposit Return Control

Current collection process

Value Damage Leakage Competing supply chain parties Collection points Consumer Responsiveness (e.g. attitude Customer Information (e.g. lack of knowledge) Multiple parties End-user (e.g. individual vs. bulk use

(possible) influencing factors

U n ce rta in ty Salesmen

(21)

21 If there is only one customer that the RTP is sent to, tracking of this task may be easy. However, if a customer is also receiving RTP from other competing supply chain partners, there is a very real chance that the RTP may be kept or stolen throughout the chain (Rogers and Tibben-Lembke, 1999). Moreover, leakage involve customer responsiveness or damage (Fleischmann, 2001). RTP could be damaged and will not be returned in the collection process. Even when customers are part of the same corporation, the customer may not have any incentive or motivation to help the supplier get the RTP back (Rogers and Tibben-Lembke, 1999).

This latter is investigated by two researchers Jonathan Fisher and Paul Horton (1979). They did a study of consumers‟ and retailers‟ attitudes and practices concerning RGB. The authors highlighted the importance of the trippage of a RGB to the current beverage container debate. Trippage is defined as the number of times that a RGB is used for filling and delivering beverages. A review was done of the then current available data on trippage, and results of surveys were presented which was undertaken in the UK, of consumers‟ and retailers‟ attitudes to RGB. In this study it turns out that retailers place restrictions on the types of RGB that they will accept from the consumer. Two main reasons given by these retailers for having restrictions were „that their supplier would not accept the RGB‟, and „the RGB caused problems for them‟ such as handling and sorting of the RGB. Thereby this survey study showed that the consumer did not return RGB because they could not be bothered or were unwilling to return RGB, or found it difficult to return RGB. These difficulties arose from an apparent lack of knowledge among consumers that the RGB was a returnable, with a refundable deposit on it, and to where they could return this RGB (Fisher and Horton, 1979).

The attribute return is the party who takes care of delivering the RTP back to their storages. This attribute involves the aspect of the combination level. That is different parties who take care of RTP could make use of different services how to class the flow of goods. The salesmen of PBL are responsible for taking back the RTP from the customer.

Moreover, the reverse logistics and thereby the return handling environments are characterized by a high level of uncertainty which includes the variety, quality, quantity, and timing of the returning products (Beulens, 2004; de Brito, 2003; Poles and Cheong, 2009).

(22)

22 In general, timing and quantity of used products that are available for reuse are determined by the former user (e.g. customer or consumer) rather than by the recoverer's (e.g. PBL) requirements. Reliable planning of collection may therefore be a difficult task.

2.3 Empirical analysis

As described in section 2.2.1 of this study, different factors are found that could have influence on capture rate of the collection process. By describing the current collection process at PBL in more detail the (possible) influencing factor(s) could be identified. The decisions that are made in the current collection process will be evaluated according to the key activities listed in the introduction phase of this section. By reviewing (e.g. measure and asses) the current collection process in terms of new process objectives and new process attributes, different problems could be addressed that could influence the capture rate.

According to Davenport (1993), process objectives should meet a number of established criteria. For example, the level of change targeted should be radical. This usually means greater than 50% improvement. As discussed in the introduction phase of chapter 1, the process objective is to do a research for finding the problems and their root cause(s) of the collection process and deal with these outcomes in order to improve the collection process. By doing this, more RTP is coming back to the company and less investment has to be done to replace the lost RTP. In terms of quantitative process objectives this would be:

- Increase the incoming RTP by 50%;

- Reduce the investment costs for RTP by 50%.

However unlike the achievement of these goals, any certain improvement for the collection process that can be achieved is a progress (or change for the better). In this research the quantitative process objective is to improve the collection process towards a capture rate of 98%.

In the literature study, different attributes of managing the collection process are discussed: - The deposit system process, for example, might specify that the process will act as an

inducement to compensate the consumer for the inconvenience of returning the RTP. But it could also specify that the process will encourage customers to get their RTP back from the consumer.

- The return system process, for example, might specify that the process increase or decrease the empowerment of the salesmen. The salesmen are responsible for taking back the RTP from the consumer. The more empowerment, the greater the responsibility.

(23)

23 This evaluation verifies the conceptual model, which leads to an empirical causal model (ECM). This ECM clarifies the problems and their root causes of the collection problem. To gather knowledge for the empirical analysis sub-questions needs to be made. These sub-questions are established from the factors in the conceptual model and the relationships between these factors (Cooper & Schindler, 2006). The sub-questions are as following:

- How does the current collection process look like with respect to the aspects of the functions and design?

- Which problem(s) and their cause(s) can be identified regarding the (possible) influencing factors?

- Which problem(s) and their cause(s) can be dealt with on short-term in order to improve the collection process?

2.3.1 Describing of the current collection process

As discussed it is important to understand an existing process before designing a new one. Often existing processes have never been described or even viewed as processes. This section describes the current state of the collection process with respect to the aspects of functions and design according to Beullens et. al (2004).

Collection infrastructure

The collection infrastructure relates to the points at which customers‟ handover the used products to the network (Beullens et. al 2002). PBL uses different types of RTP for distributing beverages. The main types of RTP are kegs, bottles, and crates.

The customers are the Managed Distribution Centre (MDC) and retailers. The retailers are divided by channels. The channels are; canteen, convenience, horeca, key retail, OTC, prestige hotel, prestige restaurant, and superette. Appendix B gives an overview of the different types of retailers per channel. In total 7810 retailers makes use of RTP that have to be collected.

The customers who buy the beverages, store full RTP at their distribution centers or retail outlets. A retailer could also choose to buy from a MDC. The MDC is responsible for returning the RTP. The consumer buys full beverages in retail outlets and returns RTP at the same or another retail location. The RTP is collected on the premises of the retailer or MDC by PBL.

(24)

24 Package Full crate Empty crate Bottles only

650ML/1000ML Rs 300,00 Rs 156,00 Rs 12,00 300ML/330ML Rs 300,00 Rs 156,00 Rs 6,00 TABLE 1.DEPOSIT PRICES EFFECTIVE AS FROM 20.11.10

Beverage Industry Retailer Consumer

Pay deposit to retailer for each returned RTP

Purchase beverages: Pay deposit

Pay deposit for returned container RTP Purchase beverages:

Pay deposit

Product flow Deposit payment

FIGURE 7. OPERATION SCHEME OF THE DEPOSIT SYSTEM AND PRODUCT FLOW

Collection policy

The collection policy specifies the moments at which a collection point (i.e. at the retailer or MDC) is serviced and the volume per collected visit. At PBL the collection point will be visited on periodic schedules. Periodic schedules are visits to a particular collection point that are planned regularly according to a fixed frequency (Beullens et. al, 2002). The flow of PBL is controlled by a basic principle, the so-called one-to-one principle. This means, when a certain amount of RTP is taken to a store, at least the same amount of empty RTP should be taken out of the store.

In order to manage this distribution operation PBL makes use of information technology. The salesmen keeps track of the monetary transactions RTP movement via a handheld called bMobile. The bMobile is linked with a system program called BASIS. This BASIS receives all the transaction and RTP movements from the bMobile.

Combination level and vehicle type

(25)

25 2.3.2 Reviewing the current collection process

This section will review the current collection process by measuring the process in terms of (new) process objectives and assess in terms of (new) process attributes with respect to the (possible) influencing factors. By doing this, the problems and their causes can be identified. Reviewing the current collection process verifies the conceptual model which results in an ECM.

To gather knowledge for analyzing (some of) the influencing (possible) factors with respect to the process attributes concerning the collection process of RTP, a survey study has been undertaken at retailers and MDC‟s. A survey study is a data gathering and analysis approach in which respondents answer questions or respond to statements that were developed in advance to gain insight into people or problems under study (Kasunic, 2005).

The objective of this survey study was to find out why some RTP are not returned, especially emphasis on RGB, how they give the empty RGB‟s in the crates back to PBL. This latter involves the handling and sorting of the RGB in the crates. As discussed in the conceptual analysis, retailers place restrictions on the types of RGB that they will accept from the consumer. Reasoning, that it causes problems for the retailers such as handling and sorting of the RGB. At this moment, PBL struggles with the corporation of the customer to help PBL to sort the RGB in the respective crate. This could (possible) contribute to the effectiveness of the collection process.

This aspect of why some RGB‟s are not returned has been addressed by Fisher and Horton (1979). They did a study of consumers and retailers attitudes and practices concerning RGB. This earlier research is used as base for the developing of the questionnaire items. This survey study consisted of the following parts:

- A survey under 367 retailers (i.e. Canteen, Convenience, Horeca, key retail, OTC, Prestige Hotel, Prestige Restaurant, Superette) with in total 13 questions;

- A survey under 10 MDC‟s with in total 12 questions.

The questionnaires were completed by 324 retailers and by 8 MDC‟s and the surveyors then personally visited all these retailers. This enabled them to give useful clarification and further elaboration to their replies. A response rate of 88% was achieved at the retailers and 80% at the MDC‟s. For further details and explanations of the survey study and results see appendix C.

Control system

The process attribute system control is influenced by the amount of retailers that makes use of RTP, the loan system, RTP acceptance of retailers and keeping RTP in stock in the winter period.

(26)

26 exploitation of existing resources and the increase of complexity (de Brito, 2003). This could result in a lower capture rate of RTP from the retailer.

As can be seen in figure 2, a decreasing performance line from October 2010 until December 2010 occur. This typical seasonal fluctuation is better known as the peak season. The demand of beverages arises in the months November and December. However, the brewery and warehouse distribution cannot meet the demand to produce and deliver the beverages. To overcome this problem, full RTP is sold on basis of a loan system4. This means that customers could have beverages up in front before the peak season starts. This loan system starts around October. By doing this, the demand can be distributed over the months and the brewery and warehouse distribution can meet the demand. Afterwards, extra trips needs to be done to take all the RTP back from the customer. This explains the high peaks in the figure after the month December.

In this light, the supply chain is not fully closed. The basic one-to-one principle becomes obsolete which makes it harder to control the flow and assure availability of RTP. This has a negative influence on the capture rate of the collection process.

The survey revealed that some retailers place certain restrictions on RTP that they will accept from the consumer. The retailers were asked if they accept all of the RGB types that are brought by the consumer, or only certain ones. 75% of the retailers place certain restrictions on the type of RGB that they will accept back from consumers. It appeared to be that these restrictions depends on;

- The type of the consumer with respect to familiarity; - If the consumer bought the RGB at their place or not, or; - Accepts only those RGB that the latter sells.

The retailers were then asked if they accept every amount of RGB that are brought by the consumer, or only a certain amount. 47% of the retailers accept only a certain amount. Most of the retailers replied that it depends on how much space, stock or number of crates they have left. The latter is a result of the one-to-one principle. The retailer will not accept more RTP when it has the same amount of upcoming delivery by PBL. Moreover, some retailers accept only from usual consumers (familiarity) and how much the latter sells. These placed restrictions could be concerning as it affects the capture rate of the collection process. Presumably this could make the consumer give up returning the RGB (e.g. by throwing it away).

4

There are two types of loan systems: (1) giving out the RTP on loan basis and the customers only paying for the beverages content, (2) giving out the RTP and the beverage content on loan basis. The second loan system loan depends on how much volume the customer always orders and if they are capable to pay this order. A customer that makes use of this loan system will have a specific code that exactly gives the value of the loan (i.e. amount of RTP and beverage content on loan). Each time when a customer gives back the RTP, the credit account will be less.

(27)

27 Another implication that the survey revealed is that the retailers keep more empty RTP in stock during the winter period (i.e. from the month July to September). 74% of the retailers do not keep more RTP during the winter period. However, 26% of the retailers keep more RTP in stock so that it possesses enough RTP in the peak season in order to meet the demand of the consumers. The more empty RTP in stock the more RTP they can get from PBL, which is a result of the one-to-one principle. By keeping more empty RTP in stock, less RTP are coming back to the warehouse. This results in a decreasing capture rate.

Deposit system

The survey is used to assess the process attribute system deposit. The retailers were asked how they currently mark the prices of the beverages they sell in RGB. The survey showed that 9% of the retailers do not indicate existence or level of any deposit on the RGB and do not charge any deposit on the RGB. Presumably, this could affect the consumer‟s perception of the returnable nature of the bottle and the level of the deposit on the bottle. Both process attribute systems deposit and control would be influenced and thereby the capture rate.

But, it seems to be that the use of a deposit is not the reason that consumers bring their RGB back to the retailer. 61% of the retailers marked the beverage prices by the method of „net‟ price excluding the deposit. The retailers replied different explanations for not including the deposit:

- The retailers do not add deposit because most of consumers do return the bottles back; - The retailers are avoiding the deposit because it is too much for the consumers or; - That the consumers are not willing to pay for it.

As mentioned earlier, the deposit should act as an inducement to compensate the consumer for the inconvenience of RGB (Fisher and Horton, 1979). In this case, is it is discussable if the deposit works as an inducement to compensate for the inconvenience.

(28)

28 Return system

The process attribute system return could be influenced by the salesmen. The salesmen is responsible for taking back the RTP. By giving the salesmen empowerment, the salesmen have to make own decisions how to check or assess the incoming RTP from the customer. That is to control the flow and prevent leakage. In this light, the salesmen could influence the capture rate.

Another subject of interest in the survey was about the way how they give the empty RTP back to PBL with respect to the condition of the crates. The RGB needs to be sorted in their respective crate. The survey revealed that the salesmen do not accept the RTP when the RGB are not properly separated in the respective crates. This means, that the salesmen can accept the RTP or not. This influence the capture rate of the collection process.

Therefore it is important to understand the customer their reason on how they give the RTP back (i.e. condition). The survey study showed that most of the customers do sort the empty RGB in their respective crates. This means the soft drinks in the red crate and the alcoholic drinks in the blue crates. Most of the customers find it easy to sort out the empty RGB in the crates. However, some retailers and MDC‟s do not have time to do sorting (i.e. time consuming) or find it difficult to sort due to the many types of RGB.

(29)

29 2.3.5 Empirical Causal Model

From the results of the empirical analysis an empirical causal model (ECM) is made. This model clarifies the actual problems and their root causes that are identified in the conceptual model, see figure 8. This model can be used to identify short-term improvements in the collection process. The (possible) influencing factors that were not identified in the collection process were:

- Damage;

- Competing supply chain parties and; - End-user.

First of all, it could be complicated and confusing for PBL to deal with multiple parties with almost a total of 7900 retailers. This could make the collection process difficult and hard to control with respect to the collection policy. By making it easier to control the RTP flow, the collection policy can be improved. This could be done by reducing the amount of retailers that makes use of RTP. However, the amount of retailers has direct influence on the collection infrastructure and cannot be improved on a short-term by managing one of the attributes. Furthermore, it would not be reasonable to reduce the amount of retailers that makes use of RTP.

Another problem that influence the control of RTP flow is assuring the availability of RTP. The main root cause lies in the loan system that is used as solution for the capacity problems at the production and warehouse distribution. By using the loan system the one-to-one principle becomes obsolete. This makes it harder to control the flow.

The survey that is undertaken by the retailers and MDC‟s identified a lot of problems that have influence on the collection process and verified some factors in the conceptual model. The factor customer (i.e. retailer) seems to have reasonable influence on the collection process. The responsiveness (e.g. attitude) of the retailer gives problems with respect to RTP acceptance and keeping stock of RTP during the winter period. Some of the retailers places restrictions on the types and amount of RTP that they will accept from the consumer. This could influence the capture rate. These restrictions made by the customer should be avoided in order to increase the acceptance and thereby the capture rate of the collection process. Moreover, the retailers keeping more RTP during the winter period due to the one-to-one principle. This influences the performance which result in a lower capture rate.

Thereby the retailer has influence on the consumers‟ perception of RTP. The retailers that mark the (gross) price of the beverage including the deposit could presumably affect the consumer’s perception of the returnable nature of the RGB and the level of the deposit on the RGB. This could affect the return rate of the RGB at the retailer. These retailers should make use of a different marking system to avoid this perception. This could improve the return rate and thereby the capture rate.

(30)

30 pints/bottles) and that it is too time consuming. This could result in a lower RTP acceptance from the consumer by for example the retailer.

Thereby the salesmen do not accept the RTP when the bottles are not properly separated in the respective crate. By making it easier and less time consuming to sort out the crate, retailers will do better sorting. This could increase the efficiency of handling and more acceptances by the salesmen. Thereby it could encourage the retailers to accept more RTP from the consumer.

Another problem is the value of the deposit price. The deposit price is presumably too high for the consumer which does not attract the consumer as before and discourage sales. An „optimum deposit price‟ should be used that encourage the consumer to bring back the RGB, does not discourage sales and still cover the cost for lost RGB.

Overall, most of the problems and their causes are covered by the factor „customer‟ in the ECM model, and especially the retailer. Thereby the customer has indirect influence or is influenced by salesmen and consumer.

(31)

31

Problems Causes Aspects of functions & designs Attributes of managing

Performance Collection infrastructure Combination level Collection policy Deposit Return Control

Current collection process

Value Leakage Collection points Consumer Responsiveness (e.g. attitude) Customer Information (e.g. lack of knowledge Multiple parties Amount of retailers making use of RTP Influencing factors U n ce rta in ty Marking system of retailers RTP acceptance of retailers Keeping stock of RTP in winterperiod Consumers‟ perception of RTP One-to-one policy (peak time) Time (e.g. consuming) Handling (e.g. space) Availability of RTP Loan system Deposit too high Salesmen RTP not sorted RTP acceptance Discourage sales

(32)

32

3. Design phase

In the diagnostic phase the current collection process is described and reviewed with respect to different problems and their root causes that have influence on the capture rate. To deal with these problems and their root causes the existing process should be improved. To design and implement a new process in an organization the following key activities should be covered (Davenport, 1993).

- Brainstorm design alternatives;

- Asses feasibility, risk, and benefit of design alternatives and select the preferred process design;

- Prototype the new process design; - Develop a migration strategy;

- Implement new organizational structures and systems.

This section discusses the first two key activities. The other key activities will be discussed in the change phase. In order to improve the collection process different design alternatives (further referred as potential projects) are described that could deal with one or more root cause(s). These potential projects are evaluated and prioritized (i.e. asses feasibility, risk, and benefit) which can be used for further research in the change phase.

3.1 Potential projects

(33)

33 3.1.1 Project 1: Differential deposit refund system

A differential deposit refund system (DDRS) makes use of a deposit and a handling fee. The intention of using a handling fee is to compensate the retailer for loss of space (storage requirements) and time (in processing the deposit and taking back the RTP and sorting) (Hogg, D et. al, 2010). By doing this the problem RTP acceptance with the root causes time and handling could be dealt with. The retailer could be encouraged to do better sorting which result in a higher RTP acceptance by the salesmen. Thereby, it could be a measure to encourage retailers to accept more RTP from the consumer (Fisher and Horton, 1979).

3.1.2 Project 2: Redemption center

A redemption center is an organization or a small business that accepts empty RTP for redemption from the public (i.e. retailer) and pay the refund deposit value. A law in the New York states defines a redemption centre as, “any person offering to pay the refund value of an empty beverage container to a redeemer, or any person who contracts with one or more dealers or distributors to collect, sort and obtain the refund value and handling fee of empty beverage containers for, or on behalf of, such dealer or distributor”5

.

A redemption center could make agreements with at least one retailer to provide them the handling and sorting services for the RTP. There are two common examples, (1) a redemption center that collects the empty RTP from a retailer, or (2) a redemption center that accepts empty RTP that are delivered from a retailer to the redemption center6.

If a redemption center collects empty RTP or a retailer delivers empty RTP, the retailer will not have the inconvenience of handling, sorting or loss of space and time. By doing this the problem RTP acceptance with the root causes time and handling could be dealt with. This could encourage retailers to accept more RTP from the consumer. Thereby a redemption center could do better sorting which results in a higher acceptance of the RTP by salesmen.

3.1.3 Project 3: Depot center

A depot center is a system whereby consumers and/or retailers take empty RTP back to a stand-alone depot for refunds (Snow et al., 2007). This depot center can be owned and operated by PBL. This depot center will function as a separate refund collection system. A depot center allows consumers and retailers to refund deposits at all time. Moreover, the depot center takes over the handling, sorting and loss of space and time from the retailers. By doing this the problem RTP acceptance with the root causes time and handling could be dealt with. This could encourage retailers to accept more RTP from

5_{http://www.dec.ny.gov/chemical/59639.html}

(34)

34 the consumer. Since this depot center could be owned by PBL, the salesmen would accept the RTP at all time.

3.2 Project choice

All of the potential projects could deal with one or more root causes but need all a disproportionate amount of time and resources to execute. This section analyses the feasibility of potential projects with respect to their relative benefits, costs, risks, and time frames. The results of these analyses provide the basis for selecting a preferred potential project for in the change phase. This potential project has the best feasibility to deal with the causes into an improved collection process.

For analyzing the feasibility of the potential projects a project selection matrix will be used. This method evaluates each potential project with respect to the relevant project criteria. The purpose of the project selection matrix is to prioritize by defining what is important on how each of the projects accomplishes goals or objectives7.

3.2.1 Project evaluation

The objective of a project evaluation is that it will be judicious, efficient and optimal use of scarce resources, avoidance of wastages and ensuring their on schedule completion with respect to criteria8. A lot of criteria in the literature can be found to evaluate and select a project. In order to select systematic useful criteria an „Affinity Diagram‟ will be used, see appendix F. An affinity diagram is a tool that gathers a large amount of language data (i.e. different project criteria) and organizes them into groupings based on their natural relationships (Brassard, 1989). In this research the project evaluation and choice will be determined by the following criteria:

- Direct links to strategic goals; - Probability of implementation; - Associate satisfaction impact; - Complexity;

- Likelihood of success.

Each criterion is assigned with a relative weight to the overall objective. At PBL the criteria effectiveness, easy to implement, and cost of implementation are used to evaluate a project. The criteria that relates to the criteria used by PBL are therefore assigned with a higher relative weight. In appendix G a table of scoring criteria for the project selection matrix can be found. Table 2 shows the result of the project selection matrix. This table shows that the DDRS project has the highest overall score and will be selected for further research in the change phase. The explanation of the criteria and impact rating of the project selection can be found in appendix H.

(35)

(36)

36

4. Change phase

In the previous section the potential projects are evaluated. The DDRS project is preferred to be selected for further research. The DDRS project will be „prototyped‟ as the new process design for improving the collection process. Developing a prototype is a way to simulate and test the operation of a new process. Prototyping must be viewed as a learning activity and helps to reduce the implementation risks (Davenport, 1993). Before prototyping this process design, the following questions need to be answered:

- How does the DDRS system works? - How to fund the handling fee? - How to value the handling fee?

- How to simulate and test the operation of the DDRS?

The product of this research section could be used for the implementation of the DDRS project. PBL has to decide whether or not to adapt the DDRS proposal in a plan of action by developing a migration strategy and implement a new organization structures and system.

4.1 Operation of the DDRS system

Within the DDRS, various stakeholders are involved in the product flow of beverages (i.e. RTP) and in the processing of deposit and handling fee payment. The stakeholders that are involved in the DDRS are:

- The beverage industry that „fill‟ the cases;

- Any retailer which sells beverages in Mauritius of PBL; - All consumers who purchase beverages of PBL in Mauritius.

An overview of the key elements, deposit -and handling fee payment and product flow in the DDRS, is given in figure 9.

(37)

37

Beverage Industry Retailer Consumer

Pay deposit to retailer for each returned container

Pay handling fee to retailer

Purchase beverages: Pay deposit

Pay deposit for returned container (e.g. RGB and/or crate) Purchase beverages:

Pay deposit

Product flow Deposit payment Handling fee payment

FIGURE 9. OPERATION SCHEME OF THE DIFFERENTIAL DEPOSIT REFUND SYSTEM

4.2 Handling fee

The handling fee will be used to compensate the retailer for loss of space and time processing the deposits and taking back the RTP. This handling fee should be funded. This section discusses how the handling fee can be funded and how the value can be determined.

4.2.1 Funding of the handling fee

The handling fee can be funded by using of two methods. Theoretically the handling fee can be included in the wholesale price to the retailers. The retailers base the retail price (minimal price increase) they charge the consumer on the wholesale price they pay to PBL for the product which consumers barely should notice9. However, pricing strategies vary between companies and are not necessarily entirely based on cost structures (Howells, 2005)

Another method to fund the handling fee is the use of the net gain as result of the unclaimed deposits. When the consumer returns the RTP to a retailer, the deposits are refunded. When a consumer chooses not to return a RTP, the deposit money is considered as „unclaimed‟. The total deposit price for a RTP is Rs 300, see table 1. On average the purchase price is Rs 250 (i.e. crate + pint/bottles). Since the capture rate of PBL is not 100%, the unclaimed deposits result in a net gain in this deposit system. This net gain can be used to fund the handling fee (Hogg, D et. al, 2010).

If the handling fee is not included in the wholesale price or can be fully absorbed by the unclaimed deposits, they may be partially or fully funded by PBL.

(38)

38 4.2.2 Value of the handling fee

There are different methods that could be used to determine the value of the handling fee. First of all, the value of the handling fee can be determined by the use of „activity-based costing‟ (ABC). ABC is a costing model that identifies activities in an organization and assigns the cost of each activity with resources to services according to the actual consumption by each (Atkinson, 2007). In this case, the services reflects the work of handling, sorting and time required for the RTP. The handling could involve the pickup, unloading and collecting of the RTP.

A more common method for determining the value of the handling fee is by the use of a fixed percentage of the deposit value. For example, table 3 shows the deposit price and handling fee used in various states of America. This table shows that the handling fee could vary between 20-60% of the deposit value.

State Deposit price Handling fees Connecticut Minimum 5 cents 1,5 cents (30%) Delaware 5 cents 1,0 cents (20%)

Iowa 5 cents 1,0 cents (20%)

Maine 5 cents 3,0 cents (60%)

Mass. 5 cents 2,25 cents (45%)

New York 5 cents 1,5 cents (30%) Vermont 5 cents 3,0 cents (60%)

TABLE 3. STAGE BEVERAGE CONTAINER DEPOSIT-REFUND SYSTEM, DERIVED FROM BOHM

(1981)

4.2.3 Discussion

As discussed there are two methods for funding the handling fee: 1. Included in the wholesale price or;

2. Funded by the unclaimed deposits

There are different commentaries saying that the net cost to an industry should be absorbed by the industry as a product cost and not passed on the consumer (Snow, 2007). This means, that the handling fee should not be included in the wholesale price. Furthermore, the results of the survey revealed that some of the retailers avoid the deposit, as it is already too much for the consumer. In this light, charging the consumer a minimal price increase on the retail price will not be encouraged by the retailer or consumer. A better alternative would be to use of the unclaimed deposits. The use of unclaimed deposits will not harm anyone and avoid comments or misbehavior of retailers.

(39)

39 average revenue that a retailer earns for selling a full RTP is Rs 43.5. This means, that a retailer earn more money with handling and sorting of the RTP than selling of the RTP. This makes no sense.

A more fair and reasonable method is the use of the ABC. This involves a deep understanding of all the activities that takes place by the retailers. However, it is hard to measure and determine all the activities that take place by the retailer. But, the internal cost of handing and sorting of the RTP is known. This would presumably be a reasonable value for the handling fee to compensate the inconvenience of handling, sorting and time by the retailer. Table 4 shows the internal costs of sorting. On average the sorting cost per RTP is Rs 0,40.

TABLE 4.INTERNAL COSTS OF SORTING

4.3 ‘What-if’ analysis

In the change phase the design solution is implemented. This means that the current situation is actually converted into the desired situation. However, a new process design should be prototyped (i.e. simulated and tested) before it can be implemented or converted (Davenport, 1993). A method that can be used to prototype and reduce the implementation risks is the „What-if‟ analysis. By analyzing of scenarios various „what-if‟ questions can be addressed. Scenarios can be defined as descriptions of fundamentally different future states of an organization‟s environment considering possible developments of relevant interdependent factors (Brauers and Weber, 1988). It is by studying scenarios that managers become better prepared to make informed decisions that take possible future developments into account (Bunn and Salo, 1993). The steps in a scenario analysis are:

- Determination of factors the scenarios will be build around. Focus on two or three most critical factors that will determine the value of the asset and build scenarios around these factors;

- Determining the number of scenarios to analyze each factors; - Estimation of the asset under each scenario.

Internal sorting

costs

Employee

Fixed salary Rs 175.000 per/annum

Annum 300 days

Rs 583 per/day

Handling

Sorting rate 55 bts/min

Costs Rs 0,02 per/bts

Pints (x24) Rs 0,53 per/RTP

Bottles (x12) Rs 0,27 per/RTP