Developing supply chain configurations for a public
library network: A case study at Biblionet
Groningen.
Written by:
Mark Wolters
S1918737
University of Groningen
Faculty of Economics and Business
Master Technology and Operations Management
June 2014
Supervised by:
Preface
This Master thesis project is the final step in completing my Master of Science in Technology and Operations Management at the University of Groningen. For this thesis I did research on data that was available from Biblionet Groningen. This dataset was the standard data from Biblionet where several people at the university already work with. By means of this dataset I tried to develop and compare different supply chain configurations. Whereas the available data was not feasible enough for this research, I requested for additional data from Biblionet Groningen. Fortunately I received this data, which enables me to continue my research on supply chain configurations within the field of public library networks.
I would like to thank several people for their help during this project. First of all I would like to thank Mr. Jos Bokhorst, who is the coordinator for MSC TOM and lecturer of the course Research Proposal. Thanks to him I gained knowledge of the requirements for writing a research proposal. Secondly, I want to thank my fellow students for their useful feedback during the presentation sessions. Furthermore, I would like to thank Mr. Tom Steffens, who acted as contact person between the university and Biblionet, and Mr. Gerlach van de Heide, who explained the dataset to me. Special thanks go out to my supervisors, Mr. Hessel Visser and Mr. Kees Jan Roodbergen, who gave me very useful feedback during the project and supported me in finishing this project.
Finally, I would like to thank Lieke de Veer, who is staff member at Biblionet. She provided me with the annual report and required data for the data analysis. Furthermore, she was always available for questions. Without her help and support it was not possible to finalize this project.
Mark Wolters
Abstract
Table of Content
Preface ... 2 Abstract ... 3 1. Introduction ... 5 1.2 Research Questions ... 7 2. Theoretical background ... 82.1 The Supply chain ... 8
2.1.1 Members of the supply chain ... 8
2.1.2 Structural dimensions ... 9
2.1.3 Supply chain integration ... 10
2.1.4. Selecting a type of supply chain ... 10
2.2 Configurations of a supply chain ... 11
3. Methodology ... 13
3.1 Stepwise approach ... 13
3.1.1 Part 1: Literature study ... 13
3.1.2 Part 2: Analysis of current practice ... 13
3.1.3. Part 3: Theory building and compare outcomes. ... 14
3.2 Case study ... 14 3.3 Data collection ... 14 3.3.1Digital data ... 14 3.3.2 Annual report ... 15 3.3 Data analysis ... 15 4. Results ... 16
4.1 Characterizing the public library network ... 16
4.1.1 Rental systems ... 16 4.1.2 Customers ... 17 4.1.3 Demand ... 18 4.1.4 Distribution ... 19 4.1.5 The product ... 19 4.1.6 Costs ... 19 4.1.5 Inventory ... 20
4.2 Making supply chains work for public library networks ... 21
4.2.2 Uncertain demand versus certain demand ... 26
4.2.3 Facility, inventory and transportation costs ... 26
4.2.4. Product availability and product variety ... 29
4.2.6 Efficient versus responsive supply chain ... 30
5. Discussion ... 32
6. Conclusion... 34
1. Introduction
There is a serious reason to think that collections and content which libraries need to manage are still increasing due to the fact of the recent information explosion witnessed by librarians worldwide. Collection sizes of libraries will still increase and will even more than double within our lifetimes considering that half of our printed books are released in the last 30 years(Agee & Naper, 2007). However, throughout the history of libraries, there has been concern about having adequate space to handle all these collections (Block, 2000). Therefore libraries across the world are seeking for storage facilities as shelves keep overflowing current capacities. For example in the USA a lot of academic libraries already have an off-site storage facility for their printed collections.
However, according to Goulding (2009) public libraries are always dependent on decisions of local authorities and the government. Therefore funding from these authorities and government is always an issue for public libraries. Due to economic influences, governments cut on funding public library networks, which means that the budgets of public library networks are decreasing (Goulding, 2009). Whereas budgets of public library networks are decreasing, public library networks should seek for solutions to decrease costs.
In business environments supply chain configurations are introduced as an opportunity to reduce costs (Chopra, 2003;Fisher & Hammond, 1997). In literature different supply chain configurations are mentioned. The configuration theory establishes patterns or profiles which describes an organization as a set of interrelated activities (Flynn, Huo & Zhao, 2010). So the term supply chain configurations refers to patterns or profiles which describe the supply chain as a set of interrelated activities. Lambert and Cooper (2000) mention the horizontal structure, the vertical structure and the horizontal position of a focal company as the different supply chain configurations.
among all libraries in the network (Stevens , 1977). So, this means that within a public library network the structure is like a horizontally integrated supply chain. There is already some literature about horizontal integration of storage facilities within a public library network. However, the concept of a vertical integrated storage facility within a public library network is still not investigated. So, different kinds of supply chain configurations are still not investigated for the public library network. Therefore the goal of this paper is defined as:
Perform a multi-criterion comparison of different supply chain configurations for public library networks.
Due to the fact that there is still little known about this subject, the aim of the paper is to build theory based on the findings. This theory will be based on literature and insights from practice. To get insights from practice a case study will be performed at the Biblionet Groningen. Biblionet Groningen is a large network of public libraries within the region of Groningen. This network consist of circa sixty public libraries, some service points and makes use of vans to distribute the books across the network. Biblionet Groningen is one of the public library networks which has to deal with decreasing budgets and wants to research the possibility of a storage facility.
1.2 Research Questions
Once the problem context is obvious and the goal of the paper is clear, the research question can be defined. The main research question of this paper is:
What supply chain configurations should be used for public library networks to decrease costs?
According to this main research question of the paper there are the following sub questions.
- What are the different supply chain configurations and what are the characteristics of these configurations?
- Which aspects are of importance by selecting a supply chain configuration? - How can a public library network be characterized?
- What are the most important performance indicators in a public library network? - How to make supply chain configurations work for a public library network?
- How are the different supply chain configurations related to the performance indicators?
For the last two sub-questions a case study is needed to get the answer. The case study will analyze the current situation and characteristics of the network in order to develop supply chain configurations, which match with these characteristics. After the development of these supply chain configurations, these configurations will be evaluated by the different performance indicators.
2. Theoretical background
In this section a literature review will be presented, which is closely related to the sub questions mentioned in the past section. The aim of this literature review is to give insights from literature on the topics addressed in the sub questions. In this way there will be a better understanding of these topics. Because the public library network will be described by analyzing the current situation, the literature review will be limited to supply chains and supply chain configurations.
2.1 The Supply chain
In literature a lot of different definitions are given for supply chains. In this thesis the supply chain will be defined according to the definition provided by Christopher (1998). This article defines a supply chain as a network of organizations, which are involved through upstream and downstream linkages, in the different processes and activities that produce value in the form of products and services in the hands of the ultimate consumer. In other words, a supply chain consists of two or more separated organizations, being linked by material, information and financial flows. These organizations may be firms producing parts, components and end products, logistic service providers and even the customer himself (Christopher, 1998). According to Lambert and Cooper (2000) the three primary aspects that are important for a company’s supply chain network structure are the members of the supply chain, the structural dimensions of the network and the different types of process links across the supply chain. Each of these aspects will be explained now.
2.1.1 Members of the supply chain
2.1.2 Structural dimensions
According to Lambert and Cooper (2000) there are three structural configurations of a supply chain network, which are critical to describe and manage the supply chain. The article mentions the horizontal structure, the vertical structure and the horizontal position of the focal company as the different structural dimensions of a supply chain. Whereas the horizontal structure refers to the number of tiers across the supply chain, the vertical structure refers to the number of suppliers and customers represented within each tier (Lambert and Cooper, 2000). A company can have a narrow vertical structure, with few companies at each tier level, or a wide vertical structure with many suppliers and/or customers at each tier level and the supply chain can be very long, with numerous tiers, or short with a few tiers (Lambert and Cooper, 2000). Figure 2.1 gives a schematic representation of these two structural dimensions.
The horizontal position is the third and last structural dimension mentioned by Lambert and Cooper. (2000). The article describes this structural dimension as the position of the company within the supply chain, which can be near the initial source of supply, near to the ultimate customer, or somewhere between these end points of the supply chain. From literature it is obvious that companies can use very different kind of combinations of these three structural variables.
2.1.3 Supply chain integration
Flynn, Huo, & Zhao (2010) define the different types of supply chain in order of integration. The article defines integration as the unified control of a number of successive or similar economic or especially industrially processes formerly carried on independently. To use this in the supply chain context, Flynn et al. (2010) define the term supply chain integration as the degree to which a manufacturer strategically collaborates with its supply chain partners and collaboratively manages intra- and inter-organization processes.
According to Flynn et al. (2010) there are three dimensions of supply chain integration: customer, supplier and internal integration. The article defines customer integration and supplier integration as forms of external integration, which is the degree to which an
organization is involved with its external partners to structure inter-organizational strategies, practices and processes into collaborative, synchronized processes. In this case customer integration involves core competencies derived from coordination with critical customers, whereas supplier integration involves core competencies derived from coordination with critical suppliers (Flynn et al. 2010). In contrast to external integration, internal integration involves activities within an organization, which is defined by Flynn et al. (2010) as the degree to which a manufacturer structures its own organizational strategies, practices and processes into collaborative, synchronized processes.
2.1.4. Selecting a type of supply chain
According to Fisher, Day and Ryan (1997) the first step in devising an effective supply chain is to consider the nature of the demand of the products a company sells. Factors that need to be considered are like product life cycle, demand
predictability, product variety and market standards for lead times and service times. Fisher (1997) distinguishes products based on their demand pattern in two categories: functional products and innovative products, which each require a different kind of supply chain. So to prevent a mismatch between product and supply chain configurations, it is
necessary to consider the nature of demand. Figure 2.2: Characteristics of functional and innovative products.
Fisher (1997) mentions two different types of supply chains: the efficient supply chain and the responsive supply chain. Whereas as the
efficient supply chain is designed to deliver products at low cost, the responsive supply chain is designed for speed of response (Fisher, 1997). As mentioned before the selection of one of those types of supply chains depends on the nature of the product offered. Figure 2.2, which is adapted from Lee (2002), describes the characteristics of functional and innovative products.
Based on these table companies could identify which type of products they produce. Figure 2.3 shows which type of supply chain is feasible for the different type of products. So, when a company has identified its type of product, the company could select a type of supply chain
that matches with the type of product. From figure 2.3 it is obvious that functional products
match with efficient supply chains and innovative products with responsive supply chains.
2.2 Configurations of a supply chain
According to Flynn et al. (2010) the configuration theory establishes patterns or profiles which describes an organization as a set of interrelated activities. The configuration theory described by Miller (1986) states that the alignment of strategy and systems or practices is reflected in the patterns observed in practice. This means that there is a need to construct patterns of supply integration. Whereas different organizations may place differing degrees of emphasis on the individual dimensions of supply chain integration, there should exist different configuration patterns of supply chain integration. These patterns can be described in terms of their supply chain integration strength and balance (Flynn et al., 2010). The different patterns derived by the configuration theory determine the relationship between supply chain integration and operational performance (Flynn et al, 2010). So this means that performance of an organization is dependent on the configurations of the supply chain of that organization. According to Dong (2001) and Truong & Azadivar (2005) there are two categories of configuration decisions on supply chain networks. The first category consists of structural decisions, which include location of factory, warehouse and retailer, capacity for each facility and transportation modes. The second category consists of coordination decisions, which
include inventory deployment, centralized or decentralized control for replenishment decisions, make-to-stock or make-to-order production policy and transshipment policies, demand forecasts and allocation rules for insufficient stocks. Chopra (2003) describes different configurations of supply chains based on the distribution network. Chopra (2003) gives six different types of supply chain configurations based on the design of the distribution network. By selecting one of these supply chain configurations, the characteristics of the distribution network should be taken into account (Chopra, 2003).
3. Methodology
In this section will be explained how the research will be prepared and executed in order to answer the main research question of this paper. In this section a stepwise approach will be presented to execute the research. Also, the case is introduced in this section and the method that will be applied for data analysis.
3.1 Stepwise approach
The sub questions can be divided in to three parts and so on the research will be done in three steps, which are crucially to answer the main research question. Table 3.1 shows how the sub questions are divided in the three main research steps.
Part 1: literature study
- What are the different supply chain configurations and what are the characteristics of these configurations?
- Which aspects are of importance by selecting a supply chain configuration? Part 2: Analysis of current practice
- How can a public library network be characterized?
- What are the most important performance factors in a public library network? Part 3: Theory building and compare outcomes
- How to make supply chain configurations work for the library sector?
- How are the different supply chain configurations related to performance factors
Now the research is divided in three main steps, a stepwise approach can be presented that makes clear how each step of the research will be executed.
3.1.1 Part 1: Literature study
The first step to take in this research is an extensive literature study. This literature study will focus on supply chain configurations in order to answer the first two sub questions. This method seems to be feasible enough to answer these sub questions, because validation can be established by comparing different sources of evidence with each other (Karlsson, 2009). 3.1.2 Part 2: Analysis of current practice
Due to the fact that there is still little know about a public library network in literature, a literature study is not feasible to answer the sub questions related to this subject. Therefore a case study is selected to do this part of the research. According to Karlsson (2009) a case study lends itself to early, exploratory research where the variables are still unknown and the
phenomenon not at all understood. So a case study is appropriate to use in this situation due to the fact that this field of research is pretty new and explorative. A single case study is
selected, because this gives a greater opportunity for depth of observations (Karlsson, 2009). 3.1.3. Part 3: Theory building and compare outcomes.
In The last part of the research, supply chain configurations will be developed according to the characteristics determined in the second part of the research. So, in the third part of the research part one and part two are linked together. This will result in different supply chain configurations, which then can be compared to each other and evaluated by some
performance indicators. By making use of the identified performance factors within a public library network, propositions about the relationship between the different supply chain configurations and performance indicators can be developed. To develop these propositions a single case study will be performed. Hak and Dul (2008) state that a proposition that
expresses a deterministic relation implies that for each single case instance in the domain the proposition is true according to the theory, which means that the proposition can be tested in a single instance. Because an experiment is not feasible for this research, a single case study is the best strategy (Hak and Dul, 2008). So, a single case study will be used to help developing the propositions. The developed supply chain configurations will thus be based on a single case study, which is feasible enough. The next section will briefly describe this case study.
3.2 Case study
The case study will be performed at the Biblionet Groningen, which is a public library
network in the region of Groningen. This network consist of circa sixty public libraries, some service points and makes use of vans to distribute the books across the network. Biblionet has to deal with decreasing budgets and therefore seeks for opportunities to reduce costs.
Furthermore, Biblionet wants to investigate the possibility to integrate a storage facility into the network to store their printed collections and books. However, these books should still circulate within the library network of Biblionet.
3.3 Data collection
In this section the data collection of the case study is described. Two methods of data collection will be mentioned.
3.3.1Digital data
which all titles of the collection are mentioned with year of release and the specific barcodes of a book. Beside data of books also other products, like DVD’s, magazines and paperbacks are included in this dataset. However, these products need to be filtered out of the dataset. Whereas this thesis only focuses on books, these other products are irrelevant for this research. So, there is already a lot of data available from this case. However, not all data is usable or feasible for this research. Therefore it is of great importance that the correct data is selected to do the research. In section 3.3, there will be explained which data is used for analysis in this thesis and what kind of methods are used to analyze these digital dataset. 3.3.2 Annual report
Next to the digital data available, also information is collected by use of the annual report of Biblionet of the year 2012. The annual report describes the current situation of Biblionet and describes important strategic decisions of the moment. Furthermore, this annual report is further explained by employees of Biblionet to get a better understanding of the information provided. With the information in this annual report and the explanations of employees it is possible to characterize the public library network.
3.3 Data analysis
data with the corresponding ISBN numbers for only one week of 2012. However, by matching the barcodes to their ISBN numbers in the rental dataset it is possible to do demand analysis. This analysis is performed and presented in the results section.
4. Results
In this section the answers on the sub questions will be presented. Whereas the different supply chain configurations are already mentioned in the literature section, this will not be done again. This section will therefore start with characterizing the public library network. According to this characterization of the network, supply chain configurations can be selected from literature and developed for the public library network. At the end of this section a multi-criteria comparison of these supply chain configurations will be performed.
4.1 Characterizing the public library network
Due to the fact that there is little literature available about public library networks, an analysis of current practice is necessary to characterize this kind of network. To do so, the annual report of Biblionet is used and several people that are related to the public library network or have knowledge of this kind of network are asked for further explanations. According to this information the public library can be characterized in terms that are also relevant in describing companies. Also characteristics of comparable businesses can be found in literature. So beside the interviews the public library network could be characterized by comparable businesses like video rental stores for example.
4.1.1 Rental systems
4.1.2 Customers
A public library has a lot of different kinds of customers, which results in a broad and very diverse customer base. In this part this customer base will be defined and elaborated. Customers can be placed in groups according to age, preferences or to borrowing behavior. A public library deals with all ages of customers. It is a fact that customers of different ages prefer different types of books, for example children will prefer the children’s books and adults prefer real novels. The second distinction between customers is their preferences. Each customer prefers different types of books. The last distinction that can be made between customers is their borrowing behavior. Some customers make a reservation for books they want to borrow while other customers just visit the library, which are known as the walk-in customers, and select their borrowings from the books that are available at that moment. Another kind of borrowing behavior is the number of books a customer borrows a time. Some customers borrow the maximum number of books a time, while others borrow just one or two books a time. Figure 4.1 gives a representation of the different type of customers. This figure is based on two dimensions, which are number of books and number of times a customer visits a library.
high rarely customers top customers
number of books
low initial customers medium customers low visit frequency high
This figure shows that it is possible to identify four types of customers along the two dimensions. However, from practice it is obvious that the upper right dimension and the bottom left dimension are the most likely types of customers in a public library. So, the focus in this thesis will be on these two dimensions. Furthermore, from practice it is known that public library networks want their initial customers to be top customers. So, this means that public library networks want their customers to be loyal. To do so, customer retention should increase and public libraries should meet customer requirements.
4.1.3 Demand
According to Fisher et al. (1997) the first step in devising an effective supply chain is to consider the nature of the demand of the products. So demand is an important factor in describing the network. In public libraries demand is very unpredictable due to the fact that there is a great diversity in customers as explained before. However, the main problem of demand in rental systems is that a substantial proportion of rental requests are not due to reservation. This proportion of demand is related to walk-in customers, which makes this proportion of demand very uncertain(Fink & Reiners 2006).
To define demand, the borrowing behavior of customers should be taken into account. Some customers make a reservation for books they want to borrow while other customers just visit the library, which are known as the walk-in customers, and select their borrowings from the available books. The demand of the walk-in customers may be uncertain, but this is not an issue in this case due to the fact that these customers select their borrowings from the available books. So, the demand of these customers is always fulfilled even if this demand is uncertain. However, there are also walk-in customers, who visit the library for specific titles. If these titles are available there is no problem and the walk-in customer can be served. But if these specific titles are not available the customer will make a reservation for these specific titles or leave disappointed. According to the borrowing behavior of the customers, demand in the public library sector can be defined as the moment when a reservation for a book at a specific facility location is made. To fulfill demand the specific titles should be available at the specific facility location.
customers who had exhausted the current items of interest. However, demand is still uncertain until a reservation is placed.
4.1.4 Distribution
Due to interlibrary loan, which is explained by the car rental example in section 4.1.4, the public library network acts as a distribution network. In the current situation, the public library network is designed as different pick up locations that exchange products with each other. According to Chopra (2003) there are different possible designs of distribution networks and within one network it is possible to have different designs. Chopra (2003) states that the most important performance factors to evaluate a distribution network are meeting customer needs and the costs of meeting customer needs. This is of high relevance for a public library network. The network has one collection, which circulates within the network. Customers can place online orders, in which they can decide which books they want to borrow at a certain moment and at a certain facility location. So, for the public library network it’s off high importance to fulfill these orders. This means that the network should guarantee that the reserved books are at the right facility within the network at the right moment. Furthermore, costs are also very important for a public library network due to decreasing budgets.
4.1.5 The product
Beside books libraries also have DVD’s, video’s, paperbacks, documents and e-books. However, books are the main products in the library network and in this paper only the books will be taken into account. As there are a lot of different kinds of books and titles, there is a high variety of products. Furthermore, there are more differences between the products. There are very popular books and less popular books. So, the public library network has to deal with high demand products and low demand products. There is also a difference in number of copies of each book.
4.1.6 Costs
information. So, if public libraries want to operate at minimum costs, public libraries should seek for solutions on those four aspects.
4.1.5 Inventory
The inventory of a public library network consists of one collection of books. This collection is stored in a lot of different library facilities. In order to get insight in the inventory of a public library network, it is necessary to know how this inventory is composed. Therefore a Pareto analysis of the inventory is made from the available data of the case company. This analysis is shown in the figure below.
From the Pareto analysis it is obvious that 25% of the titles does 80% of the inventory and 75% of the titles only represent 20% of inventory. So, this means that the inventory exists of high volume products and low volume products. These products will be identified later on in this thesis.
This paragraph characterized the public library network, which gives a better understanding of the network. The following section will present how supply chain configurations could work for this kind of network.
0,00% 20,00% 40,00% 60,00% 80,00% 100,00% 120,00% 0% 5% 10% 15% 20% %25 30% 35% 40% 45% %50 55% 60% 65% 70% 75% 80% 85% 90% 95% percentage of total number of items
percentage of total number of titles
Pareto analysis Inventory
4.2 Making supply chains work for public library networks
To make supply chains work for public library networks it is necessary that the supply chain configurations are in line with the characteristics of the public library network. To do so, criteria can be set according to the characteristics of the network mentioned in the previous section. These criteria are shown in the following list.
- Uncertain demand and more certain demand. - High demand and low demand products - Product variety
- Operate at minimum costs - Distribution network
- Customer needs that are met/product availability
Furthermore, it is clear that there are different kinds of structural and coordination decisions on supply chain configurations. To make supply chains work for public library networks, these structural and coordination decisions on supply chain configurations should match with the characteristics of the public library network. This means that these structural and coordination decisions should be based and related to the list of criteria presented above. To do so, different supply chain configurations will be compared and evaluated by these criteria in this section.
Criteria Centralized inventory Decentralized inventory
High demand products Unsuitable Suitable
Low demand products Suitable Unsuitable
Uncertain demand Suitable Unsuitable
Certain demand Unsuitable Suitable
Facility costs Low High
Inventory costs Low High
Transportation costs High Low
Product availability High Low
Product variety High Low
Response time Long Short
To determine which kind of configurations public library networks should adapt to improve their performances, all the evaluation criteria that are mentioned in the table above will be critically discussed by using the case company in the following sections.
4.2.1 High demand versus low demand products
Due to the fact that public library networks face the complexity of characteristics of both functional and innovative products, it is necessary to identify those products. To do so, a Pareto analysis is carried out. This analysis is based on the rentals during the first week of 2012. To recognize demand patterns, it is necessary to make a link between barcodes of the books and the ISBN numbers of the books. In this way it is possible to create demand patterns for the different titles. The result of the Pareto analysis, which is based on the ISBN numbers of the books, is presented in the following graph.
From this graph, it is obvious that fifty percent of the rented titles do eighty percent of the total rentals during this week. The other fifty percent of the rented titles are just good for twenty percent of the total rentals. According to these results, two different groups of products could be identified. The first fifty percent are the high demand products, which are the titles that are frequently rented during this period, while the other fifty percent are low demand products, which are the titles that are only rented a few times during this period.
Now it is proven that within a public library network there are high demand and low demand products. So, the next step is to identify which products are the high demand products and which titles are the low demand products. Therefore it is necessary to make a link between barcodes of books and the ISBN numbers of books.
To do so, table 4.2 will describe the characteristics of the ten products with the highest demand and the ten products with lowest demand within the period.
0 0,2 0,4 0,6 0,8 1 1,2 0% 5% 10% 15% 20% %25 30% 35% 40% 45% %50 55% 60% 65% 70% 75% 80% 85% 90% 95% percentage of rentals percentage of titles
pareto analysis of rentals
The product characteristics described in the table give some obvious results. There are two obvious differences in product characteristics between the high demand products and low demand products. The first difference is in the moment of release. Whereas the high demand products have in common that they are recently released, the low demand products have in common that they are already in the network for many years. Another obvious difference is the number of copies. In the case of high demand products there are many copies, while in the case of low demand products there is only one or a small number of copies available. Taking into account the result of the Pareto analysis of the inventory, the conclusion can be made that the high demand products take more inventory than the low demand products.
Based on the two differences found between high demand and low demand products, public libraries networks could categorize their books according to two dimensions. These dimensions are recency and volume. Whereas recency is based on the release date of a book and the volume is the number of copies of a title in the collection.
isbn number release date number of copies number of rentals first week of january 2012
9789020530711 2-1-2012 65 38 9789022999875 9-6-2011 99 32 9789041416537 30-9-2010 100 30 9789025858377 10-10-2011 114 29 9789085920854 5-3-2010 97 27 9789045111179 3-9-2010 116 27 9789460680496 7-11-2011 48 27 9789021805498 6-10-2011 75 27 9789085921301 4-2-2011 84 26 9789041418487 30-6-2011 92 25 9034519538 24-8-2005 4 1 9067935832 20-9-2001 2 1 9034519783 18-8-2005 2 1 9067936308 13-4-2004 2 1 9034732339 8-3-2007 1 1 9067936456 26-1-2005 2 1 9034747352 3-1-2007 1 1 9067936847 9-1-2004 3 1 9022529967 14-2-2002 1 1 2707317918 14-11-2002 1 1
high the classics top books of the moment
volume
outdated books slow movers
low
not recent recency recent
From this figure it is obvious that books can be categorized into four categories. These categories will now be explained in more detail.
Outdated books: these are the books that are not recent, but released years ago. Furthermore,
there are only some copies of these books available in the collection of the library network. Demand for this category of books is lowest, so this category consists of the low demand products.
The classics: this category of books is not released recently, but there are more copies of these
books in the collection available than is the case for outdated books. This category of books is also more popular than the outdated books. However, demand for these books is still low in contrast to the books recently released.
Top books of the moment: this category consists of books that are recently released and there
are a lot of copies available in the collection of the library network. The demand for these books is highest, so this category of books consists of the high demand products.
Slow movers: this category of books consists of recently released books, but there are only a
few copies available in the collection of the network. Demand for these books is much lower than demand for the top books of the moment.
The division of books in these categories makes it possible for public libraries to identify the composition of their collection. To do so, public libraries should identify the link between ISBN numbers and barcodes, which makes it possible to do analysis on demand patterns. Based on these categorizations a public library network could make supply chain decisions for every category. Now the different product categories are identified, it is possible to evaluate the different supply chain configurations for these product categories. In the demand analysis
only the top books of the moment and the outdated books are identified at the moment, so in the following sections of this thesis only these two categories will be taken into account. According to table 4.2 above the high demand products should then be stored decentralized and the low demand products in a centralized inventory. However, it is not just that easy to determine the design of the distribution network. Public library networks should not only take the criteria of high and low demand products into account, but also the other criteria mentioned in the table. These criteria will be covered in the following sections.
4.2.2 Uncertain demand versus certain demand
In the section before the distinction is made between high demand products and low demand products. Furthermore, a method was presented to divide books in four different categories. Due to this method it is possible for public libraries to identify their high demand and low demand products. By taking the identified demand patterns into account it is obvious that demand for the top books of the moment is more predictable and reliable than demand for books the outdated books. So, demand for top books of the moment is some kind of certain and demand for the outdated books is quite uncertain. Whereas decentralized inventory is unsuitable for uncertain demand and centralized inventory is unsuitable for certain demand, it is logical to use decentralized inventory for the top books of the moment and centralized inventory for the outdated books. The next section will discuss the costs of such kind of inventory strategy.
4.2.3 Facility, inventory and transportation costs
In this section the costs of the different configurations will be critically discussed. First of all the costs of the current configuration, which is total decentralization, will be analyzed. Secondly, the proposed configuration of the section before will be analyzed. Because costs are the most important performance measure for public library networks also costs of one alternative configuration, which is total centralization, will be discussed to compare this with the current and proposed configurations.
facility when a reservation is made. In this situation inventory is stored at a lot of different local facilities. According to Selim & Ozkarahan (2006) the main advantage of this design is that it lowers transportation costs and the main disadvantage is the high facility and inventory costs. However, according to Biblionet’s annual report of 2012 the number of reservations increases with 18%, which means that there is need for more transport for reservations and so on transport costs will increase. Furthermore, Selim & Ozkarahan (2006) state that this design of the distribution network is only suitable for fast-moving items where customers value high response rates due to the fact that slow moving items are increasing inventory costs.
Taken these facts into account the option of storing the fast-moving items in local facilities and the slow-moving items in a centralized facility would be a better option. In this way inventory costs would be decreased, because the slow moving items are stored in a central warehouse. However, due to the fact that the fast moving items are still stored decentralized, facility costs will still be even high as in the current situation. Furthermore, there is still transport needed for books that are returned to another facility than where they are rented and for reservations of the slow moving items. Nevertheless, it is possible for this situation to decrease facility costs, because due to the fact that only fast moving items are stored decentralized the inventory level at the local facilities is much lower. The public library network could therefore think about decreasing the number of facilities by integrating different small inventory facilities into one large inventory facility. To do so, the model of Chopra (2003), which is presented in figure 4.2, could be used. This figure indicates that a certain number of local facility locations in the network will decrease the transportation costs as long as inbound transportation of economies of scale are maintained (Chopra 2003).
According to this, the public library network should seek for a situation in which there is a balance between facility costs and transportation costs to keep total logistic costs low. This transportation costs will mainly consist of the return of books to the right facility and delivery of reservations. However, by integrating different local facilities in some bigger facilities will decrease the costs of these kinds of transport. Furthermore, the transport costs of reservations and returns for the slow moving items, which are stored in the central warehouse, will be low. From the rental analysis in section 4.2.1 it is obvious that for these items there is only one request per week. So, a possibility could be to do transport only once a week for these items, which will reduce costs.
From this section it is obvious that the current situation is the most expensive one. The proposed inventory strategy, the combination of decentralized inventory and centralized inventory, could be costs saving if the correct number of facilities is chosen and integration of facilities is applied. However, the situation of total centralization seems to be the cheapest solution. Beside this comparison on costs for the three situations, the next sections will deal with some other performance comparisons of the three situations.
4.2.4. Product availability and product variety
Whereas the previous section critically discussed the performance measure cost, product availability is another performance measure for public library networks. Product availability is the probability of having a product in stock when a customer order arrives (Chopra, 2003). So, this means that customer requirements should be met. In paragraph 4.1.2 are the different customers identified and based on this it is obvious that there is a high diversity of customer requirements. Furthermore, from section 4.1.2 it is known that public libraries want the initial customers to become top customers. Therefore it is necessary for public libraries to meet a lot of different customer requirements. To do so, public libraries should offer a high level of product variety to meet the different preferences of their customers. Also, Public libraries should guarantee a high level of product availability to their customers to meet their requirements. According to the evaluation criteria in table 4.1 the current situation of total decentralization is not feasible to guarantee high product variety and high product availability. However, it is necessary to notice here that customers that visit the library just choose from the available books or otherwise place a reservation for books. And also the product variety at each facility is pretty high due to the high number of different titles available at each facility. So, in that case the total decentralized situation is feasible, because customer requirements are almost always met in this case.
requirements of most customers are met. Nevertheless, the availability and variety of products at the decentralized facilities is low, which makes it harder for public libraries to get the initial customers to be top customers. Furthermore, reducing the number of facilities by integration could also lead to a decrease of customer retention, because from the annual report of Biblionet it is obvious that the number of visiting customers decreased when some facilities are closed.
From the inventory analysis in paragraph 4.1.5 it is obvious that public library networks have to deal with a high variety of products due to the high number of different titles. Table 4.1 states that centralized inventory are much more suitable for high product variety and high product availability environments than decentralized inventory. So, it would mean that total centralization would be the best configuration for the public library network. However, in this case the customer experience of visiting a library is totally disappeared and from Biblionet’s annual report of 2012 it is obvious that they are losing customers due to closing some facility locations or change a facility location into a service point. So, by taking away the experience of visiting a library it would make it much harder to attract customers and even harder to get customers from the category initial customer to top customers. Furthermore, there is even a certain possibility of losing customers.
From this section the conclusion can be made that public libraries should make a tradeoff between the kind of configuration that is most suitable for their network and the loyalty of their customers. In the case of more suitable configurations, it is more difficult to stimulate customers to use the library service more frequently. Now the different configurations are evaluated on costs, product availability and product variety. The next section will present another performance measure for evaluation of the different configurations.
4.2.6 Efficient versus responsive supply chain
varying degrees of functional and innovative characteristics in the real world. Furthermore, these characteristics are not frozen in time, because as products mature through their product life cycles the customer requirements changes (Aitken et al. 2003). This is also the case for public library networks. Based on recency and volume books are classified in one of the four categories. However, recency and volume of titles in the collection will change over time. This means that it is necessary to do continuous review on recency and volume to get the right titles in the right category. Due to the fact that library products have a mix of characteristics of functional and innovative products it is not feasible to take all characteristics of these products into account. Whereas recency is identified as the most important factor for demand, the choice between functional or innovative product is based on the selling season. Whereas only the top books and the outdated books are identified by analysis at this moment, only these two categories will be discussed in this section.
From the Pareto analysis of rentals it is obvious that rentals of books are highest at the moment they are first released. These are the so called top books of the moment. So, this means that there is a kind of short high selling season for these products. So, these top books of the moment could be identified as innovative products. As mentioned in the literature section innovative products match with responsive supply chains. Whereas table 4.1 states that decentralized inventory have a high response time, a decentralized configuration fits best for the top books of the moment.
The category outdated books has a much lower number of rentals according to the Pareto analysis. This category of books acts more as books with a long product life, because there are still a minimum number of rentals. The outdated books could then be identified as the functional products. From the literature section it is clear that functional products ask for an efficient supply chain. This means that response times for these products are higher and distribution happens more efficient. According to table 4.1, which states that centralized inventory has high response times, a centralized configuration would be better for the outdated books.
the preferred balance or should make a tradeoff between these two performance indicators. The different configurations are now evaluated on all performance measures stated in table 4.1, so the next section, which is the discussion section, will give a total evaluation of the outcomes.
5. Discussion
In this section, the results of the previous section will be stepwise presented in sequence of the sub research questions mentioned in section 1.2. In this way, it will be obvious how the results are related to the sub questions proposed. Furthermore, these results will be critically discussed and limitations of the research will be discussed.
This thesis characterized the public library network as a distribution network. Furthermore, this distribution network is characterized by high and low demand products, a rental system, high and low volume products and high variety of products. The thesis also identified some performance factors for this network. These performance factors are operating at minimum costs, increase customer retention, responsiveness and meet customer requirements. However, this characterization is only based on one case. This can make it difficult to generalize the results for all public library networks. Furthermore, it is known that operating at minimize costs is the most important performance factor for this public library network, but it is unknown how important customer retention, responsiveness and meeting customer requirements are. Also no data on these performance factors was available for research, which made it impossible to make these performance factors measurable.
Whereas literature describes a lot of different kinds of supply chain configurations, which varies from configurations based on structural and coordination decision to configurations based on integration determined by performance factors, this thesis only takes into account
configurations based on the design of a distribution network. This is a result of the fact that this thesis characterizes the public library as a distribution network and only focuses on this factor of supply chain configurations. Whereas the design of a distribution network is focused on decentralization versus centralization, this thesis only evaluated the case of a totally decentralized network, a totally centralized network and a combination of these two options to develop supply chain configurations. To develop more complete supply chain configurations for public library networks a lot of other factors, for example facility location, transportation modes or capacity of facilities should be taken into account.
To make supply chain configurations work for public library networks, this thesis used and identified characteristics of the public library network and evaluation criteria. The main characteristic of the network that is considered is demand, because according to literature the nature of demand is the key driver to select supply chain configurations. To determine the nature of demand a data analysis in the form of a Pareto analysis is performed. This gave the opportunity to classify books into different categories along the dimensions of recency and volume. Classifying books into these categories makes it possible to make decisions about supply chain configurations for the different categories. The result of this is that a centralized configuration is better for outdated books and a decentralized configuration is better for the top books of the moment. However, in the data analysis in this thesis are only two of the four categories identified, which are the outdated books and top books of the moment. To make the analysis more complete, the other two categories should also be identified. In this way also more complete decisions on supply chain configurations could be developed. Furthermore, this data analysis is only based on data of the first week of 2012. Due to the large amount of data in the dataset, applying the Pareto analysis for all the periods in 2012 was not achievable within the time scope of the research. To get more reliable results it would be wise to do the analysis for all the periods during 2012. The evaluation criteria were used to evaluate the suitability of the different configurations. However, these evaluation criteria are only based on one article. Furthermore, it is unsure if all evaluation criteria are of same importance or if there is a difference in importance between those criteria. So, further research on these evaluation criteria is necessary to check for validity and differences in importance.
decentralized inventory and centralized inventory, could be costs saving if the correct number of facilities is chosen and integration of facilities is applied and the situation of total centralization seems to be the cheapest solution. However, due to the fact that there was no data available on costs, these outcomes cannot be proven. Cost analysis based on modeling each of the three situations with real input data, are necessary to make this result more valid and reliable. Furthermore, the configuration with decentralized and centralized inventory fits best for responsiveness and the product characteristics of the network, while the configuration of total centralization scores worst on customer retention and meeting customer requirements. According to this it is obvious that not one of the configurations scores best on all performance measures.
6. Conclusion
References
Agee, J. & Naper, S., 2007. Off-site storage: an analysis. Collection Building, 26(1), pp.20– 25. Available at: http://www.emeraldinsight.com/10.1108/01604950710721566 [Accessed March 17, 2014].
Aitken, J., Childerhouse, P. & Towill, D., 2003. The impact of product life cycle on supply chain strategy. International Journal of Production Economics, 85(2), pp.127–140. Available at: http://linkinghub.elsevier.com/retrieve/pii/S0925527303001051 [Accessed June 9, 2014].
Buchan, R.P. & Siegfried, J., 1978. An economic evaluation of the magazine distribution industy. Antitrust Bulletin, 13, pp. 19-50.
Chopra, S., 2003. Designing the distribution network in a supply chain. Transportation
Research Part E: Logistics and Transportation Review, 39(2), pp.123–140. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S1366554502000443.
Christopher, M.G., 1998. Relationships and alliances: Embracing the era of network competition, in Gattorna J. (Ed.) Strategic management journal Hampshire: Gower press, pp 272-284.
Dong, M. 2001. Process modeling, performance analysis and configuration simulation in integrated supply chain network design. Doctoral Dissertation, Virginia: Polytechnic institute and state university, Blacksburg.
Dul, J. & Hak, T. 2008. Case study methodology in business research, Burlington, USA: BH Elsevier.
Fink, A. & Reiners, T., 2006. Modeling and solving the short-term car rental logistics
problem. Transportation Research Part E: Logistics and Transportation Review, 42(4), pp.272–292. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S1366554505000050 [Accessed May 9, 2014].
Fisher, M. & Hammond, J., 1997. Configuring a supply chain to reduce the cost of demand uncertainty. Production and Operations Management, 6(3), pp. 211-225 Available at: http://onlinelibrary.wiley.com/doi/10.1111/j.1937-5956.1997.tb00427.x/abstract [Accessed June 22, 2014].
Fisher, M.L., Day, G.S. & Ryan, W., What is the Right Supply Chain for Your Product?
Flynn, B.B., Huo, B. & Zhao, X., 2010. The impact of supply chain integration on performance: A contingency and configuration approach. Journal of Operations
Management, 28(1), pp.58–71. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0272696309000412 [Accessed January 21, 2014].
Goulding, a., 2009. Editorial: Credit Crunch: the Impact On Libraries. Journal of
Librarianship and Information Science, 41(1), pp.3–6. Available at:
http://lis.sagepub.com/cgi/doi/10.1177/0961000608099894 [Accessed June 22, 2014]. Karlson, C., 2009. Researching operations management, New York: Routledge.
Koschat, M.A., Berk, G. & Blatt. J., 2003. Newsvendors Newsvendor Tackle Problem the.
Interfaces, 33(3), pp.72–84.
Lambert, D.M. & Cooper, M.C., 2000. Issues in Supply Chain Management. Industrial
Marketing Management, 29(1), pp.65–83. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0019850199001133.
Lee, H.L., 2002. Aligning Supply Chain Strategies with Product Uncertainties. California
Management Review, 44(3), pp.105–119. Available at:
http://www.jstor.org/stable/info/10.2307/41166135.
Miller, D., 1986. Configuration of strategy and structure: towards a synthesis. Strategic
management journal, 7(3), pp 233-249.
Min, H. & Zhou, G., 2002. Supply chain modeling: past, present and future. Computers &
Industrial Engineering, 43(1-2), pp.231–249. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0360835202000669.
Pasternack, B.A. & Drezner, Z.V.I., 1999. The videotape rental model. Journal of Applied
Mathematics and Decisions Sciences , 3(2), pp.163–170.
Roehl, R. & Varian, H.R. 1996. Circulating libraries and video rental stores. First Monday, 6(5).
Selim, H. & Ozkarahan, I., 2006. A supply chain distribution network design model: An interactive fuzzy goal programming-based solution approach. The International Journal
of Advanced Manufacturing Technology, 36(3-4), pp.401–418. Available at:
http://link.springer.com/10.1007/s00170-006-0842-6 [Accessed May 27, 2014]. Stevens, C., 1977. Governance of library networks. Library trends, 26(2), pp. 219-240.
Truong, T.H. & Azadivar *, F., 2005. Optimal design methodologies for configuration of supply chains. International Journal of Production Research, 43(11), pp.2217–2236. Available at: http://www.tandfonline.com/doi/abs/10.1080/00207540500031998 [Accessed May 12, 2014].
