Tracking and Tracing in the Fishing Industry: Implementation & Validation of TraSer

(1)

Master Thesis

Tracking and Tracing in the Fishing Industry:

Implementation & Validation of TraSer

A case-study By Alex Fekken

University of Groningen Business Administration Msc: Operations & Supply Chains

Theme: Logistics & ICT J.C. Wortmann D.J. Van der Zee Blueport Lauwersoog

M. van Bergen

(2)

2

Abstract

This research discusses tracking and tracing in the fishing industry combined with a new perspective on the current asset management. It focusses on the platforms TraSer and Euridice for validation and implementation. The inefficiencies of the current situation and the cost of these inefficiencies will be outlined. After this analysis, the research proposes an organizational redesign and outlines necessary links between (existing) information system; concluded with a redesign of the processes and a proposition for an action plan. It will turn out that not only the packaging requires unique codes but also each batch (of fish) unloaded from the boats.

Key words: Tracking & Tracing, TraSer, Euridice, Fishing Industry, Unique Coding

Research theme: Tracking & Tracing

Supervisor: J.C. Wortmann

(3)

3

1. Introduction

This study discusses possibilities of implementing a tracking and tracing system in the fishing industry in the Netherlands. The research focusses on the costs and requirements of a tracking and tracing system in the supply chain. Clients demand more transparency regarding the location of where the fish (including shrimps) are caught and in what conditions the fish has been. Governmental institutions are demanding tracking and traceability of food-products: the "Council of the European Union defined it as the ability to trace and follow food for human consumption, feed for animal consumption, animals destined for human consumption or ingredients, through all stages of the supply chain" (Regulation 2002). It is not yet possible to provide tracking and tracing information throughout the entire supply chain from the boats through the auctions to the buyers and end-users. That is because it is difficult to invest in a complete tracking and tracing system as a single player in the fishing supply chain so it requires national adoption. It will turn out that the investment needed can perhaps be covered by a joint effort starting a pool of crates in the sector. Such pooling may reduce the investment needed and thereby provide an opportunity for nation-wide tracking and tracing.

The introduction continues with a short explanation of tracking and tracing before moving on to the objective of the research. Finally the importance of the pooling of crates in the fishing industry will be emphasized.

1.1 Tracking and Tracing

ISO (2000) states: "traceability is the ability to trace the history, application or location of that which is under consideration". Jansen (1998) makes a distinction between product tracking and product tracing. "Product tracking originates from product value or risk, whereby one wishes to locate the products. Product tracing originates from exception handling, whereby one wishes to establish the source of (bad) quality". According to Sallabi et al. (2011) a "traceability systems are record keeping systems designed to track the flow of product or product attributes through the production process or supply chain". This research defines tracking as being able to follow an object (real-time) including its attributes and tracing being able to access information on where the object originates and the route it has taken to get there.

(5)

5

1.2 Pooling of Crates

Coinciding with the research on tracking and tracing is the pooling of crates. The crates used in the industry are owned by the auction centers (individually) and are not interchangeable between auctions.

A tracking and tracing system is expensive to implement for a single organization. Furthermore it is highly impractical for the supply chain to implement multiple tracking-systems from different organizations. To cover the expenses for such a tracking and tracing system the auction centers could cooperate by using the same crates. However such a pooling-system is not in place yet and discussing tracking and tracing without it seems highly theoretical. In order to have a cost-efficient tracking and tracing system the system should be implemented throughout the supply chain, which can be done through pooling (and tracking) the crates.

2. Methodology

This research will try to validate TraSer-principles. The research will use TraSer for guidelines for the implementation of the tracking and tracing system in this case-study. Secondly this research will use results of the other project, Euridice, which focusses on goods instead of processes. It will help model the tracking and tracing system for the fishing industry. These systems were chosen as they seem suitable for the retail industry and the fresh foods industry. Furthermore both systems are open-source which makes it comparatively to makes changes to fit specific demands of the industry.

The research objective is to combine researches on tracking and tracing solutions

(TraSer and Euridice) in order to validate these and show their applicability in an inter-organizational environment in a fresh foods supply chain. The fishing industry involves small batches of production without a set geographic origin (fishing boats), this combined with changing packaging and different heterogeneous parties within the supply chain makes tracking and tracing difficult. The academic relevance of this research will be the unique identification for both the packaging (crates) and the catch (the fish). This will result in two different sets of data as explained further in the research. The research question is:

“How can tracking & tracing improve the supply chain of the fishing industry?"

2.1 Research Method

In order to answer the research question several sub questions have to be made.

• General

• What are the processes in the current supply

chain?

(6)

6

• What are the benefits of improved asset

management?

After researching the current situation the conditions and costs of a shared pool will be examined. The costs of both scenarios will be compared in order to show the (financial) benefit of a pooling-system. Lastly the implementation costs will be examined using TraSer logic. This part of the research is mostly a case-study involving interviews with different stakeholders. As the auction centers are the owners of the crates the focus will lay here.

• Tracking and Tracing

• What are the requirements for the tracking

and tracing system?

This will be the start of the second part of the research. Every tracking and tracing system is different based on needs and objectives; these needs will be researched in the theoretical framework. They will be examined for the fishing industry based on demands of the EU, buyers and fish auction centers. This information will be gained through the same enquiries and interviews.

• What are the requirements for data storage

and retrieval?

What information should be collected in order to meet the requirements (as demanded by the buyers, retailers and consumers)? How detailed should the tracing-information be? Is stating that they were caught this week, in the North Sea by a certain boat detailed enough or should it entail the exact moment and exact location. Furthermore the tags will be subjected to different temperatures, saltwater, and rough conditions. What are these conditions and will the tags be able to sustain these?

• How should the implementation of the

tracking and tracing system be done in the supply chain?

The question on how to develop the system remains. Not only the information on the tags is important but also who would own this data. Who will host the tracking-system behind all these tags and who should develop such a system? Who will cover the expenses of developing, implementing and maintaining the system?

To answer these sub questions current literature on tracking and tracing will be examined, though the focus will lie on previous case-studies on tracking and tracing projects such as TraSer.

• Implementation

• How will the revised supply chain look like? The last sub question will explain and discuss the differences in the processes that will be made in order to implement both the pooling system and the tracking and tracing system. What does the new supply chain look like, after implementation of tracking and tracing and pooling of crates? How does the industry get to that revised supply chain, what are the main problems that need to be overcome? In summation these are the sub questions:

General:

• What are the processes in the current supply chain?

What benefits can be made through improved asset management?

Tracking & Tracing:

• What are the demands for the tracking and tracing system?

(7)

7

• How should the implementation be done in the supply chain?

Implementation:

• How will the revised supply chain look like?

2.2 Scope

The scope of the research is the Dutch supply chain involved with the crates. This research will focus on the efficiency that can be gained through tracking and tracing and the pooling of crates. The emphasis will be on the platforms TraSer and Euridice.

The research itself is preliminary, which means that no statistical tests will be conducted.

Analyzing and mapping the possibilities of using TraSer and Euridice will give a clear overview of how it can benefit the supply chain. The research will focus on the business side without going into detail into the exact workings of the ICT. The focus will be on the processes that need to be changed, the benefit such a system and its implementation.

3. Theoretical Framework

The packaging will be discussed first; secondly the theoretical framework of tracking and tracing in general and finally TraSer and Euridice will be discussed specifically.

3.1 Reusable Articles

Carrasco-Gallego et al. (2012) states that the term “reusable articles (RA) refer to durable products intended to be used multiple times by different users in different locations of a supply-chain network’. This implies that the time the product is used by each consumer is short compared with the article lifetime, and

that each use cycle does not significantly deteriorate the product. It also implies that after each use, the RA needs to be returned to an adequate location where it is made available for the next user”. Saphire (2004) research showed that the purchasing cost of reusable articles is significantly higher than those of one-way containers. A reusable article, such as a plastic box, can cost 10 times more than a one-way article, such as a corrugated box. Witt (2000) stated that “due to the rather high purchasing costs of reusable packages, it is important to have a high utilization rate and a minimal loss rate for them”.

Before tracking and tracing can be implemented the reusable items (the crates) must first be pooled. Kroon and Vrijens (1995) define six attributes (issues) of a pooling system, as summarized by Kärkkäinen (2004): ownership, return, storage/maintenance, control, deposit/rent and monitoring. Ownership “shows the owner(s) of the packages in each of the rotation systems”. Return comprises of who returns the packages to the start of the supply chain. Storage and maintenance entail the responsibility for warehousing, repair and cleaning. Control means keeping count of the packages, deposits/rent whether or not to charge rent or hold deposits. And finally monitoring discusses the information management level; to what extent (individual or account-based) should the packages be tracked. Each of these attributes will be outlined in different parts of the study.

3.2 Tracking & Tracing

(8)

8

the individual returnable transport units are and in what condition they are in. This limited visibility creates the tendency for people and organizations to feel less responsible for proper management of RTIs. As a consequence, unnecessary costs resulting from high loss rates, breakages and unavailability of RTIs are generated which have a negative impact on the overall performance of the whole supply chain”. As Shamsuzzoha & Helo (2010) note: “the tracking and tracing of logistical networks is identified as the search for competitive advantage, where manufacturing firms have recognized the potential importance of their logistical networks". Kärkkäinen et al. (2004) states “tracking the movements of individual packages, instead of relying solely on accounts based on quantities enables improved control and administration”. In short improved visibility through asset management will decrease breakages and loss rates.

Golan et al. (2004) state "the characteristics of good traceability systems vary and cannot be defined without reference to the system's objectives". Different objectives help drive differences in the breadth, depth and precision of traceability systems. Breadth describes the amount of information the traceability system records. Depth of traceability system is how far back or forward the system tracks. In many cases the depth of a system is largely determined by its breadth. Precision reflects the degree of assurance with which the tracing system can pinpoint a particular food product's movement or characteristics". These are all aspects that will be taken into account while designing the framework for the fishing industry.

3.2.1 TraSer

Kelepouris (2007) states that "the emerging technology of radio frequency identification (RFID) creates great opportunities for effective and efficient traceability system design. Automated data capture enables traceability information to be obtained at significantly reduced labor costs and with small changes in the enterprise’s business processes. Furthermore, RFID technology combined with the appropriate information infrastructure can enable end-to-end traceability in the supply chain at small costs, affordable by small and medium sized enterprises". “Radio frequency identification (RFID) is an emerging technology that is increasingly being used in supply chain management. RFID technology plays an important role in supporting logistics and supply chain processes because of their ability to identify, trace and track information throughout the supply chain” (Zhu et al. 2012). As Hermosillo et al. (2009) state; “being able to get the present and historic information about a product is very important in the supply chain management, since it helps to make the right decisions at the right moment.”.

TraSer1_{is a “TRAceability SERvice” for the} Retail supply chain that uses existing RFID technology, although it would also be applicable with e.g. barcodes. Hermosillo et al. (2009) state about TraSer that it is an “RFID infrastructure that will allow each tag to be considered individually, and then add a traceability service to get the information about the product's location and environment and complement the package with persistence, so that not only the present status of the

(9)

9

product is considered, but also the historical data that has been gathered can be used”. By adding an information system to this RFID infrastructure and considering the tag individually allows the system to track an item across multiple locations and over the lifecycle of the item. Furthermore TraSer uses a notation that allows for independent namespace for the ID’s which means that these identifiers can be created independently. The database behind TraSer does not have to be restricted to a certain IT-structure and limits to the data, for instance data about the GPS of the catch, could be easily integrated with the database.

Figure 1: TraSer architecture

Figure 1 shows the system behind TraSer, it receives signals from the readers, containing the reader ID, time of scanning and the electronic product code. Step 2 is translating these signals to a business event; step 3 is storing the data into an information system to allow for quick access. Step 3’ is the system that allows external entities to access the information. Step 4 allows the generated information to be accessed and step 4’ to

create new events through the application interface

In short TraSer is an open-source tracking and tracing solution platform suitable for smaller companies; which can be tailored according to specific needs while retaining its flexibility in additional development. Focused on supporting tracking and tracing on the level of individual items (as opposed to the account-based view of purely quantitative stock levels), and relies on web services for communication with a focus on small enterprises.

3.2.2 Euridice

(10)

10

viewpoint the Euridice2_{project combines the} most advanced technologies available today to create a transparent and well-structured system that is able to handle the complexity of modern logistics scenarios.

Figure 2: Euridice, Intelligent cargo

The platform specializes in dealing with perishable goods. It enables transport with automatically tracking and tracing capabilities while gathering data as the temperature and other variables that might affect the goods. Euridice is a flexible platform that can use any kind of identification technology (like RFID); therefor it might be suitable to complement TraSer. For more information on Euridice see the first appendix.

4. Examining the Supply Chain

of the Fishing Industry

Before (potential) improvements through the pooling crates can be calculated the inefficiencies must be identified first. In order to do this the processes of the current supply

2_{Visited on 16 November 2012,} http://www.euridice-project.eu/

chain involved with the fishing crates will be described.

4.1 Information Systems

The systems currently in use are the E-logbook system on board of the boats and the auction systems of the fish auction centers. Interviews and enquiries with employees of auction centers and available information on the logbook revealed this information.

The auctioning system in use by the auctions is either EFICE or PEVA. Either one is used by auction centers in the Netherlands though both work in a similar way. Buyers can purchase fish here from anywhere using a digital auctioning clock and all buyers have a unique code to purchase on. The auctioning system has data on the buyers, the load and the boat it was delivered by.

Every boat is equipped with an E-logbook that monitors where the boat has been, where it has fished and how long it has been out to sea via GPS. This information is uploaded to a central information system maintained by the Dutch government. Even though the EU requires traceability through the E-logbook the system is not yet in contact with systems of the auctions nor the buyers.

4.2 Current Supply Chain

(11)

11

The supply chain starts when the fishermen go out to fish and load up with ice (for preserving fish) and fishing crates. The second process starts when the boats return to unload the fish at the auction center, and third and fourth processes start at the transport and at the unloading at the buyer. To further clarify the supply chain a figure of the lifecycle of a crate is shown. In order to emphasize the different interactions between the stakeholders and systems a figure of the entire supply chain of crates is shown in conclusion.

 Fishing Boat

Figure 3: Current processes fishing boats

Figure 3 shows the fishermen load up with ice and fishing crates from the auction centers before they go ‘out’ for the week; these crates are (individually) owned by the auction center. Most auction centers keep track of how many crates are sent with the fishing boats but not which ones. When the boats return

the crates are filled with the fish ready to be sorted at the auction center. During fishing the boats activate a GPS system (the E-logbook), kept on board, which the auction center can track real-time and government can trace using an information system.

 Auction Center

Figure 4 shows the process when the fish is unloaded. They are divided on size (and quality) and put into (partially) new crates (as more crates are needed after sorting); where they are ready to be weighed. By law each load needs to be tagged by using a piece of paper to keep track of what boat unloaded the fish. After weighing, the fish will be auctioned off; the details of the load are sent, either digitally or through a paper checklist, to the employee in charge of auctioning. Depending on the center this is auctioned off almost immediately or in a matter of days. After auctioning, which is done digitally using an information system, the fish will be made ready for transport and another paper will be added on top of the crates to make it identifiable for the buyer. A second paper called the ‘load-list’ is printed out; these contain all the fish that was bought by an individual buyer.

Figure 4: Current processes fish auction centers

(12)

12

which then drive to and unload at the auction centers for processing. This means the process of loading up with crates

differs slightly as well.

 Transport

After the auction centers prepare the crates for transportation (the crates are tagged with a piece of paper with information) the core-processes of the

auction center end. After the transport (a truck) arrives its driver will identify the right set of crates, using the ‘load-list’, load it on the trucks for transport to the specific buyer(s).

Figure 5: Current processes transportation

These buyers are (historically) mainly clustered around three main areas; the companies in these clusters might share the transport when possible. However also common is that a single buyer will provide her own transportation. It differs between companies and auction centers allow different kinds of methods. There is no main method in handling the transport from the auction centers to the buyers.

 Buyer

Having bought the fish and coordinated the transport the buyer now waits until these arrive. After this arrival the fish will be unloaded and be put in the processes of the buyer, most of the buyers immediately unload the fish off the crates to be processed and conserved. The crates are not used further in

the system and will go back with the next transport.

Figure 6: Current processes buyers

The process of unloading the fish differs between the buyers; different types of fish and purposes require different methods of processing. Depending on the fish different steps are required. Almost every buyer promotes traceability as one of their capabilities, though only from the start of their own processes with limited visibility of earlier processes in the supply chain.

After unloading the fish is either placed on an (partially automated) production line right away or put in storage first. Storage of plaice is done in ‘tubs’, which are large crates capable of handling bulk-goods. These ‘tubs’ will then be sorted and processed further separately so tracking and tracing is possible. Other fish might be put into crates made for 20 to 50 kilo’s for preservation (the crates used by the auction centers are made for 40 kilo).

(13)

13

 Other

Auction centers have different methods of calculating the costs of renting these crates to the buyers, the amount charged and how accurate they track these crates. It can be stated that all auction centers track the amount of crates

but not which crates. Tracking the amount is done automatically via the same system as used for the auctioning itself. Not shown in the

process description is the cleaning of the crates, when the crates return they are cleaned using a cleaning machine. Each separate auction center owns a cleaning machine and each machine has a large amount of overcapacity, as earlier studies have shown. Furthermore each crate uses the same mold so they are identical except for the color and its imprints.

After the crates are cleaned they are put in stock ready to be used again until it is established they are worn out after cleaning.

Figure 7 shows the life-cycle of a crate. When the crate deteriorates too much it will have to be thrown away. Depending on the usage this can take up anywhere between 5 to more than 20 years.

Figure 7: Lifecycle crates

(14)

14

 Total Picture

Figure 8: Current processes fishing supply chain

Figure 8 shows the entire current supply chain of the crates. Important to note is that there is very little communication between the auction centers and the buyers. On the contrary the auction centers, the depots and the boats communicate a lot already.

4.3 Inefficiencies

(15)

15

done the completeness of this list is significant.

 The fishermen do not always end up going back to the same auction center causing crates to end up at other centers. They need to be transported back by truck (or boat) which brings extra costs.

 There are situations where crates are being used by boats that belong to auction centers where they neither load nor unload. Another situation is when boats load up crates at an auction center but do not unload there: certain types of fish are no longer required by law to be auctioned.

 Buyers do not always return the crates; which means crates end up at the end of the supply chain where they are not being used. When an auction center tries to pick up these crates he is only allowed to pick up its own.

 The crates of auction center A often end up at a buyer near auction center B as there are three main concentrations of buyers. It would be more efficient to transport them to center B however this is not possible because they are owned by center A.

 This might cause these crates to remain at the buyer, where they have not been cleaned. These crates are usually placed outside under the influence of the weather conditions. These give bacteria and a chance to spread and will attract vermin to the crates. This will make it difficult to clean the crates once they are back at

the auction, having a detrimental effect on hygiene.

 Each auction center has a separate cleaning machine (for the crates) with considerate capacity; however this means that utilization is low (Thors & Duineveld 2005).

Trucks are often hired to bring back crates to the right auction center, which costs money. There are also licensed companies that search for crates in for instance in England where they have a service that returns crates to the auction centers receiving one euro per crate. These inefficiencies will be discussed and explained further in the next chapter. They exist due to a lack of asset management and a lack of cooperation in the supply chain. Asset management would increase insight in where the crates are and increased cooperation between the supply chains would mean the crates could go to the nearest center. A pooling system could increase the cooperation between the auction centers. This would allow the centers to implement an efficient tracking and tracing system together, which in turn would make asset management possible. 4.4 Cost of Inefficiency

Before the necessity of a tracking and tracing system can be argued the (potential) benefit must be outlined first.

(16)

16

fish in this industry in 2011. On average boats take about 400 crates with them to sea. This means that when all boats are fully loaded they take up 141.200 crates. The auction centers themselves estimate the amount of crates in use by the boats to be around 150.000. They estimate the amount in use by the buyers or on (external) depots to be around 90.000 and the remaining 60.000 in own depot.

The first inefficiency mentioned is that boats not always return to the same auction center when loading and unloading fish. This results in that approximately 1.600 Dutch fishing crates are in stock at any given auction center belonging to different centers (on average), this number was determined through the enquiries of different centers and regular checking. This is stock that cannot be used (unless transport back to the original auction center is done). This means that 1.600*10 (cost price of a crate) = 16.000 euros that are, unnecessarily, locked in assets per center. A total of 11 centers * 16.000 costs per center = € 176.000 locked for the Dutch supply chain. Eventually these crates will need to be transported back to the centers they belong to, which means extra costs need to be made. When these crates finally get returned after an extended period of time this usually costs an extra fee of one euro per crate for the transportation companies. As the turn-rate of these crates is about a half year and costs € 17.600 this means a direct inefficiency cost of € 35.200 per year.

A second inefficiency is the usage of assets by third parties, meaning that others (either fishing boats or others) use the fishing crates but are not renting them from the auction

center. This happens when the crates are lost and other organizations decide to use them. The third and fourth inefficiency are both inefficiencies where the crates are not used but are in depot at the buyer for whatever reason. This happens when the buyer no longer needing new fish for a period of time. As the transport is too expensive to send for a single set of crates they will be on the terrain of the buyer until it buys new fish from that auction center.

These inefficiencies are caused by a lack of control on the assets. Most centers do not know how often this occurs. However all these are caused by a lack of insight in asset management. For instance it happens often that unclean crates remain on depots even though they are near a center, which is not good for hygiene. The supply chain estimates they have to write off almost 22.000 crates yearly (in total) due to the crates no longer fulfilling quality requirements. This number will be reduced with asset management capabilities using a tracking and tracing system. These inefficiencies are caused by a lack of asset management thus they are not visible in the data.

(17)

17

right away. Less reserves are needed after the pooling the crates as the auction centers can help each other. As Qinghua et al. (2009) states in a similar situation on pallets a decrease by a third is feasible after pooling. For the fishing industry this would be 20.000 crates which is a saving of €200.000. The same would apply to the external depots where 90.000 crates are stored. Which means 30.000 crates can be saved, which is another €300.000.

The supply chain estimates that they misplace about 70.000 euro worth of crates every year (7.000 crates). Some of these end up going to foreign countries. Despite hiring organizations that return approximately 3.000

crates a year, for a fee of one euro, there are still 7.000 crates that do not return. This amounts to an extra cost of approximately €3.000 for fees of returned crates and €70.000 for the crates that remain missing.

The cleaning machines cost a lot of money, both in investment and in

yearly costs, while the usage percentage is very low. Each auction center has its own cleaning machines, even though most crates go to one of three (geographical) concentrations of buyers. Larger cleaning machines with more capacity work more efficiently. After comparing the costs of the machines they can be up to €0.05 cheaper per crate which means that if 300.000 (average of the smaller centers) times a crate is cleaned accumulate to a yearly difference of €15.000 for each (smaller) auction center.

The machines already present on each auction center do not have to be replaced. They could still clean the crates that do not ‘go near’ one

of these main hubs, utilization would just be lower.

Shown in table 1 is an estimation on cost savings that can be made. These are the costs that can be saved per year when the crates are pooled and a full tracking and tracing system is implemented. Note that this research determines the costs of the investments using the interest that could’ve been received during this time; this would on average be at least 4%. The total amount of investments that does not have to be made is: €676.000,-. However as this is an investment that has been made it is not a direct saving and is not calculated as such in the table above.

Table 1: Potential for saving

There are many other advantages that are difficult to quantify. Improved asset management will increase insight in processes which allows for other improvements in the processes. The main advantage and goal of tracking and tracing is of course being able to provide tracing; as the government requires food industries to do so by law. Furthermore stakeholders are starting to demand tracing so it can be used as a marketing feature. Euridice is able to assist coordination between the different transport organizations using information gained from TraSer. As mentioned when examining the current processes most of the organizations in charge

Potential for saving Calculation Yearly costs

Unloading different auction 4% € 176.000 € 7.040

Returning crates € 35.200

Inefficiency crates 4% € 500.000 € 20.000 Cleaning machines € 15.000 a center € 120.000

Returned foreign crates € 3.000

Loss of crates € 70.000

(18)

18

of transportation do not structurally cooperate; just on a local level. As almost every form of cooperation between transportation companies is done ad-hoc this creates inefficient transport (Thors & Duineveld 2005). They state they estimate that organizing transport more efficiently will bring a saving of € 500.000 yearly for the supply chain.

This saving can be achieved through information on required transport-movements, which is made available by TraSer. Euridice is a system that is able to coordinate cooperation of transport while making other value-added options such as temperature control possible using information gained from TraSer.

5. Implementation of Tracking

& Tracing

This chapter will elaborate on the implementation of tracking and tracing system in the fishing supply chain. These are the organizational problems that need to be overcome, linking of the systems and the access requirements. Followed with the processes, the investment costs and an action plan.

5.1 Organizational Measures

There are different organizational hurdles that need to be taken which arise with these changes. Who is in charge of cleaning machines, charging fees for the crates centrally or de-centrally?

5.1.1 Organization

These issues could be solved by creating an entity owned by a share of each auction center; this organization would own all the

crates and be in charge of maintenance. The share of each center could be relative to the amount of crates brought into the pool by the center. This is possible because each center uses identical crates (different colors). The organization would be non-profit and provide its shareholders with the crates.

There is already a similar organization for the crates used for the shrimps; the centers could decide to increase the capabilities of this organization and extend its responsibilities. The auction centers have different fees and different methods of requiring payments. TraSer is able to support the centers with individual preferences, so this does not have to be an issue.

Furthermore now that crates are no longer necessarily brought back to the auction center cleaning will become another issue. The machines can be equipped with a reader to read the tags as the crates are cleaned. It can also identify (and store) which crates are cleaned to trace it back to the auction center it was used by. This way each center will still pay proportionally for the crates used.

(19)

19

5.1.2 Access Requirements

Questions that arise in any information system are what information needs to be stored behind the codes and who should have access to what information and who is responsible for this information.

One of the advantages of TraSer is that the tags do not need to maintain a lot of information, only a unique code which leads to the information stashed on an information system behind the tags. This means data on crates can be accessed by anyone that has access to Wi-Fi and a simple reader capable of reading RFID chips. This same information can also be accessed using the information system. The newly created entity in charge of the pool would also be in charge of the data. Every user will need some degree of access; these users are the auction centers, the buyers and the retailers. As mentioned: the tag can provide the necessary information until the crates are no longer used in the supply chain. The auction centers themselves also need access; the capabilities of TraSer must be that a single fish at the retailer can be traced through the supply chain to the boat it was caught by and the (approximate) location it was caught in. Furthermore TraSer should be able to show where the other part of that ‘load’ went: to which buyers other parts of that load were sold.

5.1.3 Unique Coding

The retailers do not use the crates in the supply chain; in order to still provide the end-users with the necessary information a solution is needed.

A unique code linking to a ‘load’ or batch instead of a crate would allow end-users to examine the origin of the fish. These users would need the unique code, the ‘load-ID’ which can link up with the boat ID, a week nr, the GPS locations and the ‘load’: how the fish was caught and what fish (Plaice type 1 etc.) were caught in what quantity. Other data could be added when required by the users. If this ID is still present at the retailer, or if this information is uploaded in the systems of the buyer full tracking and tracing is possible. This unique ID could be a barcode, a digital number or a QR-code.

5.2 Linking the Systems

Links between existing information systems and TraSer (and Euridice) will be outlined here.

5.2.1 TraSer

Shown in figure 9 are the links after the implementation of TraSer. The numbers show the new information streams. The first link is the mutual information stream that will need to be implemented for the auctioning system and TraSer. TraSer will for instance need information on which buyer purchased the fish.

Figure 9: Linking the information systems

(20)

20

TraSer will keep track of which crates are on board of which boat by scanning the crates before the boat loads these. The E-logbook on board of these boats keeps track of where a boat is and where it has been. But the system also has data on the locations the boat has fished at what time and with what method. Tracing the origin of the fish, or in other words the location the fish was caught, will be possible with a connection between these two systems.

5.2.2 Euridice

It’s possible to add another link between TraSer and Euridice. Euridice is suitable for coordination of the transportation of the buyers of each auction center that is connected to the system. It is specialized in transportation of perishable goods and would allow measuring the temperature of the goods in transport. Adding this feature would allow for a complete insight in all the steps the fish take before reaching the buyer. However more expensive tags are necessary for this latter feature. These tags can potentially be installed in the cargo-area of the truck instead of on crates.

5.3 Future Supply Chain

The ‘future’ situation of the processes, after the tagging and pooling the crates, is shown below.

 Fishing Boat

In the current processes the fishermen pick up the ice and crates from the auction centers. However instead of counting the crates the tags are scanned, using Wi-Fi and the information is sent to the information system. The status (in TraSer) of these crates will be changed from being in stock to being rented

out. TraSer will be able to show per individual boat what crates are rented to which boat.

Figure 10: Future processes fishing boats

When the boat returns the crates are filled with fish ready to be sorted further on the auction center, the state of these crates will be reset having been returned. There is a possibility that in the future entirely new crates will be used after sorting (obligated by law), then they can be reset through a scanner on the cleaning machine.

The change in the processes is that the scanning of the crates is done either automatically through the gates or through manual scanning instead of counting the amount of rented crates. More information will be available on which crates were rented for what time.

 Auction Center

After the fish is sorted in (partially) new crates, the tags will be scanned to send information via Wi-Fi to TraSer informing that they are ready for auction. This way the center will know what fish will be auctioned off directly.

(21)

21

used for the tracking and tracing separately from the asset management.

Figure 11: Future processes fish auction centers

After the fish has been auctioned off the center can change the status and add information on what buyer has bought the fish for both the tags on the crates and the ‘load-ID’ on TraSer. For this a data-link with the auction system (either EFICE or PEVA) is needed.

In the meanwhile the work-floor will prepare the fish for transport. Depending on the wishes of the individual buyer the auction can still add an extra paper on the load.

 Transport

Figure 12: Future processes transportation

The transport will pick up the fish, using a reader to identify the fish or via the traditional way (piece of paper). The crates will be transported to the buyers.

When opted for gates and integrated with transport this means that the auction centers can automatically scan which crates and which load was picked up by which truck.

 Buyer

After the transport arrives the fish will be unloaded and be put in the processes of the buyer, this can be done traditionally but in order to provide the complete tracking and tracing there are alternative ways as discussed earlier (direct link of systems or QR code).

Figure 13: Future processes buyers

(22)

22

There are two difficult issues that need to be solved before successful tracing can be implemented. The first is the transfer of fish from the crates to other packaging and the second is combined processing.

If the buyer is able to track a load from the auction center to own storage TraSer can expand on this for tracing. If this is not the case a barcode could be attached to the load to give the buyer capabilities of tracing. The second issues is combined processing, if fish from different loads is processed together tracing will become less accurate. As there are many different buyers with differing processes more research is needed.

 Asset Management

The transport also takes care of the return of the crates, choosing the closest center to hand in the crates. Counting these crates is done automatically through gates or manual scanning. A small reader can be placed in each cleaning machine to reset the crate to having been cleaned and being in stock at which center.

When the crates are not returned after a set period of time TraSer can give a signal. Then the center can easily examine which crates were lost to see at which party these should be in stock and return these.

TraSer allows for tracing with multiple functionalities. Asset management and

tracking have already been discussed however tracing is also an important part. For instance when at a location spoiled (or poisonous) fish turns up TraSer can search where and how this was caused. Then TraSer will follow up with identifying other loads that passed through the same location so that these can be destroyed. Or when it turns out a certain fishing location is spoiled it is possible to track all the boats that fished at that location and identify and destroy all the ‘loads’ involved so that damage is minimized. It is therefore important to keep the data on the loads stored for a fairly long time.

 Tags

TraSer allows for separate unique coding creation, depending on the organization behind the pooling this will allow tagging the crates de-centrally or centrally.

(23)

 Total picture

Figure 14: Future processes fishing supply chain

Shown in figure 14 is the ‘future’ situation of all the stakeholders involved with the crates. Even though there are a lot of small changes and updates to TraSer most of these can be done automatically and won’t substantially change the current system.

Information transfers between the auction centers, the fishing boats and depots are already present. However the information exchange between the buyers and the centers

(24)

24

5.4 Investment Costs

The implementation costs of a tracking a tracing system will be substantial despite pooling resources. For a comparison the minimum and maximum investment of the hard-ware will be outlined below. The costs of chips, readers and gates were examined through an interview with employees from Aucxis to determine the price of quality equipment.

The total amount of fishing crates in the supply chain is estimated to be around 300.000. As the cost of a crate is about 10 euro this means the total assets in finances becomes € 3.000.000. Tagging the crates with a RFID chip costs € 1,30 per tag. This is an investment of € 390.000 when tagging the crates with a single tag.

The software validated in this research: TraSer is free-ware. There are different methods to keep track of the different crates, a reader can keep track of what crates are going

where and put this into the information system and the system will know how to process this information. Handheld readers are less expensive than handheld writers. A full tracking and tracing system would require two of these readers to use before renting the crates to the fishing boats on the auction centers and when sent to the buyers. Furthermore a reader would need to be installed in the cleaning machine to reset the information behind the tags.

All three readers need Wi-Fi access so the information system (TraSer itself) can process the data. The minimum required investment for a successful tracking and tracing system is in the table below.

If opted for a more automated system this would require ‘gates’, these can cost up to 20.000 euro per gate. The main advantage is that they can automatically ‘see’ the crates leaving the center. This can automate TraSer further but is a more expensive option. An employee does not have to manually scan these tags; however an auction center would require several of these gates for each exit. On average an auction center would require approximately three gates at a cost of € 20.000 a gate which will mean an extra investment of € 660.000. Furthermore there are a few indications that a crate might require two identical chips so that when one malfunctions the other can still provide tracking, but this will depend on the tag.

Table 2: Investment costs

The table shows the minimum investment needed for a tracking and tracing system but when opted for gates and extra chips this number can increase to € 1.605.000,-. It is clear that the investments will need a few years to create a return on investment. The implementation costs will also be considerate. 5.5 Action Plan

The processes above were described after tracking and tracing is implemented. However not discussed yet is how to implement the system. There are different steps that need to be made:

 Tagging the crates

 Pooling the crates

Investments Cost per item Amount

Tagging crates € 1,3 300.000 € 390.000

Readers € 5.000 33 € 165.000

Implementation 1

(25)

25

 Asset management o Cleaning crates

 Tracking and Tracing o Transport benefits

 Tagging the crates

First of all the crates need to be tagged before the crates can be pooled and followed. Interviews revealed that it is not practical, considering the life-time, to wait until new crates are implemented. Older crates will need to be tagged with unique codes until all 300.000 crates in the supply chain are tagged. This will cost approximately €1,30 euro for one chip per crate. The time needed for this will depend on the capabilities and amount chips in stock however it is not a difficult process.

Tagging the crates with a chip will require welding these to the exterior in such a way that it will not affect piling up the crates and will still be easy to scan. Which means either the front or back side.

 Pooling the crates

After having tagged the crates they can be pooled into a single stock for the entire supply chain as discussed earlier in organizational hurdles. This does not necessarily mean a single color has to be used however all crates are available to the auction centers; these crates will be easily identifiable due to the unique code linked to each tag.

 Asset Management

All crates are now tagged and pooled into a single set of crates for the supply chain. Before information can be used the systems of the auction centers need to be integrated with TraSer. Furthermore this also requires the implementation of the process-changes as discussed in the chapter before. The

information can then be made available to the auction centers and used to its benefit.

TraSer will be able to control the streams of crates. As the system keeps track of which stakeholders currently rent which crates; it will give a signal when crates do not return in a given period of time

TraSer enables to visibility of the crates and thus increases the control throughout the supply chain.

o Cleaning crates

A big portion of the costs can be saved on the cleaning of the crates. If these larger machines with more capacity cleaned more crates increasing utilization this could save money as discussed above in the costs of inefficiency.

 Tracking and Tracing

The last important step that needs to be undertaken is making tracking and tracing available. Before anything can happen TraSer will have to link up with the other systems. As mentioned in the paragraph on future links it would require links with both the auctioning systems and the GPS-system.

Only after TraSer has linked up with these systems can the information be bundled. This can be put this under the ‘Load-ID’ and sent further in the supply chain, making tracking and tracing possible.

o Transport benefits

(26)

26

6. Conclusion

This thesis has answered the initial sub questions and the research question. However the main research goal, as previously described, is a validation of the TraSer system. Before moving on to the academic relevance the practical relevance will be outlined first. 6.1 Practical Relevance

One of the key points of the Blueport organization, as discussed by Blueport’s Guideline (2012), is to find ways to create value instead of too much focus on cost efficiency. Both TraSer and Euridice are systems that can help with that. In this paper Blueport states the auction centers need to develop into a multifunctional room for buyers where logistics can be centralized to connect different parties in the supply chain, rather than being just a location where the fish is auctioned off.

The current supply chain provides no possibility for tracking and tracing from ‘the boat to the plate’. Despite traceability being marketed by buyers they are very limited in doing so. TraSer is a system that is capable of helping these stakeholders (the buyers) look beyond their own processes for a complete tracking and tracing. While Euridice is able visualize and coordinate more efficient transportation.

Both systems inherently require a lot of coordination and cooperation between different parties in the supply chain. The government needs to assist with the GPS system, auction centers cooperation for the pooling, transport parties’ cooperation for more efficiency and buyers would need to connect their systems to TraSer.

The estimations on the implementation costs vary between approximately € 555.000 and € 1.605.000 but will mean a yearly saving of € 255.240 for the auction centers and another € 500.000 for the supply chain as a whole. This adds up to approximately € 750.000 in potential savings per year.

Important to note is that a tracking and tracing system will add value to the product. Furthermore the EU is demanding complete traceability of all industries dealing with food.

 Risk Factors

Of course there are factors involved that work against implementing a pooling-system or tagging all the crates.

Most auction centers rent the crates cheaply only to cover minor expenses even though others might make a profit on these crates if they are not returned. Even though TraSer is able to provide different fees for auction centers and clients this might not be a sustainable system. Furthermore the significant investment necessary in a time where the amount of fish caught is lessening is a factor against implementation.

6.2 Validation

(27)

27

cooperation between different stakeholders (government, retailers).

A solution of tracking and tracing proposed in this research is the unique identification codes for both the crates and the load with decentralized creation; which TraSer is able to this. Furthermore its system of containing the information on a cloud-server accessible from anywhere with internet is highly suitable for this supply chain with many different parties. Any stakeholder at any time can look up information on origins of a crate (or load) as long as the right reader or code (load-ID) is available to the organization. The auction centers can track and trace both assets and loads using the information system.

After the implementation of TraSer costs of transportation will be less, less cleaning costs of the crates, less costs due to misplacing the crates, a slower quality decrease of these crates and improved value added processes due to offering a complete tracking and tracing system.

In short it seems TraSer is capable of handling the requests and demands of the fishing industry. This validates TraSer as a system suitable for implementation in this scenario. The tracking and tracing system as suggested will not be able to track each fish individually to a single location it was caught but to a limited amount of locations near each other. The first would require more investments (mainly on the boats); a lot of changes to the processes which would be time-expensive and stakeholders do not require information to this detail.

Euridice can be integrated with TraSer to improve the supply chain in-dependably through coordination of transport (more efficient and more insight) and temperature control (adding value).

6.3 Future Research

There are possibilities of cooperation with the fishing supply chain in Belgium as they are looking for ways of tracking and tracing as well. There are two main advantages to this. The first one are increased scale effects and the second another inefficiency in asset management. Belgium and Dutch boats sometimes load in the Netherlands and unload in Belgium (and vice versa). This means foreign crates are in stock at auction centers and inevitably Dutch crates are at Belgium centers. Research could be done in this area to see how effective it would be to cooperate between these two supply chains. Furthermore this research has shown TraSer’s validation and how it could be implemented on a business side. However the exact workings of the ICT-side have not been discussed. This study can work as a guideline for research about these ICT aspects.

Limitations

(28)

28

individual stakeholder was interviewed or sent an enquiry; through extrapolation the costs and benefit for the supply chain were examined.

The focus at the start and the extrapolation resulted in limited visibility on the heterogeneous group of buyers. More research should be done if the processes are as described for each buyer.

Finally the main specific requirement for a suitable tracking and tracing system is the separate (decentralized) code creation for tags and batches. It was not examined was if there are other systems equally suitable for this task.

Acknowledgements

Writing this thesis is not something I could have done alone. Specifically I would like to thank professor Wortmann who has helped me with quite a lot of feedback. Furthermore I would like to thank everybody at the fishing auction centre of Lauwersoog for their pleasant company and assistance with my research.

(29)

Appendix A: Euridice

Schumacher et al. (2009) state that on one hand “the possibility to control the strategic high-level processes via business related software like Business Process Engines and Service Oriented Orchestration and on the other hand it also covers most specific problem domains for single cargo items like sensing the temperature inside a truck. The presented solution maintains a clear and straight forward architecture and offers a scalable solution that is able to cope with the dimensions of today’s logistics sector”. Capabilities of intelligent cargo are shown in the figure above. Euridice has the following objectives:

 Support the interaction of individual cargo items with the surrounding environment and users on the field.

 Improve logistic performances through application of the intelligent cargo concept and technologies in the working practices of operators and industrial users.

 Develop collaborative business models to sustain, promote and develop an intelligent cargo infrastructure.

 Realize more secure and environment friendly transport chains through the adoption of intelligent cargo to support modal shift and door-to-door intermodal services.

Euridice approach in order to achieve this is:

 Adapt and integrate state-of-the-art identification, communication and networking technology (i.e. RFID tags readers/writers; GPRS localization

systems; mobile devices; software defined radio technologies) into an Intelligent Cargo Integration Framework (ICIF).

 Realize Cargo Intelligence applications for distributed and centralized data gathering, anomaly detection, analysis and prediction, based on approaches like semantic web and domain ontologies, advanced context technologies and distributed intelligent agents.



Implement interoperability standards and service oriented architectures for public and private stakeholders to access and use the services they need on a cargo item at any point along its route across European corridors, connecting the cargo with back-office users and consumers

(30)

References

Adviesgroep Blueport (2012) Een handreiking aan de regionale Blueports

Carrasco-Gallego R., Ponce-Cueto E., Dekker R. (2012): Closed-loop supply chains of reusable articles: a typology grounded on case studies, International Journal of Production Research, 50:19, 5582-5596

Golan, E., Krissof, B., Kuchler, F., Calvin, L., Nelson, K., Price, G., (2004) Traceability in the U.S. Food Supply: Economic Theory and Industry Studies. USDA Agricultural Economic Report Number, 830.

Hermosillo G., Ellart J., Seinturier L., Duchien L., (2009) A Traceability Service to Facilitate RFID Adoption in the Retail Supply Chain. Proceedings of the 3rd International Workshop on RFID Technology - Concepts, Applications, Challenges IWRT 49-58.

Ilic, A., Jason, W. P. Ng, Bowman, P., and Staake, T., (2009). The Value of RFID for RTI Management. Electronic Markets, 2009, 19(2–3):125–135.

ISO (2000), ISO 9000:2000. Quality Management and Quality Assurance – Vocabulary

Jansen, M.H., (1998). The environment as a resource: developing environmental information systems based on enterprise resource planning software. Ph.D. thesis. Eindhoven University of Technology.

Kelepouris T., Pramatari K., Doukidis G., (2007) "RFID-enabled traceability in the food supply chain", Industrial Management & Data Systems, 107:2, 183-200

Kroon, L. and Vrijens, G., (1995), “Returnable containers: and example of reverse logistics”, International Journal of Physical Distribution & Logistics Management, Vol. 25, No., 2, pp. 56-68.

Laurier W., Poels G., (2012) Track and Trace Future, Present, and Past Product and Money Flows with a Resource-Event-Agent Model, Information Systems Management, 29:2, 123-136.

Paganeli P., (2008) Brochure Euridice: EURopean Inter-Disciplinary research on Intelligent Cargo for Efficient, safe and environment friendly logistics. Italy

(31)

31

Qinghua Z, Jun W., Guoquan C., Zhuan W., Dawei W., Shanshan Z., (2009) Pallet Rental Information System Based on RFID School of Mechanical Engineering, University of Science and Technology Beijing

Regulation No (EC), 2002, 178/2002 of European Parliament 28/09/2002 for food safety issues, Formal EU Newsletter 1.2.2002 L 31/1.

Rönnqvist M., Bernstein F., Caldentey R., D’Amours S., (2012) Special Issue on supply chain management and collaborative logistics. International transactional inoperational research, 631-632 Sallabi F., Fadel M., Hussein A., Jaffar A., El Khatib H., (2011) Design and implementation of an electronic mobile poultry production documentation system. United Arab Emirates University. Computers and Electronics in Agriculture 76, 28–37.

Saphire, D., Delivering the Goods: Benefits of Reusable Shipping Containers, 1994 (INFORM, Inc.: New York, NY).

Schumacher J., Gschweidl M., Rieder, M. (2010) “EURIDICE – An enabler for intelligent cargo for the logistics sector”, Journal of Systemics, Cybernetics and Informatics (JSCI), vol. 8, no. 2, pp. 18-28. Shamsuzzoha A., H., M., (2011) Real-time Tracking and Tracing system: Potentials for the Logistics Network. Proceedings of the 2011 International Conference on Industrial Engineering and

Operations Managemen, Kuala Lumpur, Malaysia, January 22 – 24, 2011

Thors M., Duineveld M., (2005) Transportsysteem Vislogistiek: De vis laat zich uniform kisten Witt, C.E., (2000), “Are reusable containers worth the cost?”, Materials handling

management, Vol. 55, No. 7, pp. 75-80.

Xu L., D., (2011) Information architecture for supply chain quality management. International Journal of Production Research, Vol. 49, No. 1, 183–198

Tracking and Tracing in the Fishing Industry: Implementation & Validation of TraSer

Master Thesis