An agent-based integration of Flow Cost Accounting in the SCOR-model
By
Bas van der Schaaf
SIMTech supervisor: John Paul
1st Supervisor of University of Groningen: Herman Balsters 2nd Supervisor of University of Groningen: Thomas W. de Boer
S I N G A P O R E
E
G N T
H
S E S E O
C S
A F
S H C E T M S
R U
University of Groningen SIMTech
Management summary
This report is the result of my research at SIMTech. It is exploring the possibilities of integrating Flow Cost Accounting (FCA) in the SCOR-model using Agent technology. The research is the first step in the Supply-Chain Flow Cost Accounting Process Program of the SCMC. The aim of the research is defined as follows:
‘To describe a possible integration of Flow Cost Accounting techniques in the SCOR- model, and to design an agent-based software structure for this integration.’
To achieve this goal, first I have worked out a manual integration of FCA in the SCOR- model. To integrate one method in the other, you have to understand each separate method first. So first FCA and the SCOR model are described. After that I have integrated FCA in the SCOR-model. Before I get to the design of an agent-based system, a brief research in agent technology is required. Finally I have designed a framework for a FCA/SCOR system:
Flow Cost Accounting:
FCA cuts costs and relieves stress on the environment by means of an accounting approach based on the actual flow of materials. The material flow through an organization is recorded in a very detailed way using existing planning tools, often ERP systems. With this detailed information, the material flow becomes more transparent so inefficiencies in the processes can be identified, more realistic material costs can be calculated, and the material quantity needed for the processes can be reduced.
The SCOR-model:
The SCOR-model is a framework for supply-chain management. It views the supply-chain as a hierarchical structure of processes, with each process consisting of some process elements.
The top-level consists of 5 processes: plan, source, make, deliver and return. By making use of these building blocks, any organization can be described using a common set of definitions. This way, disparate industries can be linked to describe any supply-chain.
Integration:
With some adjustments, the graphical representation of the supply-chain configuration in the SCOR-model can be interpreted as a material flow model in the FCA method. This way every SCOR-process and SCOR-level can be treated like a FCA material flow model. In chapter 4 I have described a method that applies all the FCA functionalities to the SCOR-processes. In fact I have added a sixth core process to the existing five core SCOR-processes: the FCA- process. I have defined five process elements for the FCA process:
• FCA1.1: Filter data out of the ERP system
• FCA1.2: Fill out FCA card and create material flow model with data
• FCA1.3: Calculate direct material costs
• FCA1.4: Calculate system costs
• FCA1.5: Create different reports for total cost overview Agent technology:
Because the FCA database is enormous, a manual execution of the FCA process takes a lot of effort. That is why an automated system is needed. In the second stage of the research I have designed a framework for a Multi Agent System (MAS), an agent-based software tool to automate the FCA process in the SCOR-model.
A MAS is composed as a set of intelligent software agents each responsible for one or more activities in the system, and each interacting with other agents in the planning and execution of their responsibilities. The agents are cooperating with the user, and are interacting with the
ERP system. In this intelligent system the FCA-process has an extra functionality: the MAS itself is capable of analyzing the management reports the system is generating.
The design:
The MAS consists of two parts: a SCOR-foundation and a FCA module. In future more modules can be added. The SCOR-foundation contains five agents, one for each main SCOR- process. The agents are planning the production schedules, and the agents solve perturbations in the schedule by reviewing the schedule. The agents are coordinated by the Plan-agent, which is generating the schedules and communicates them to the other agents.
The FCA module consists of just one agent, the FCA-agent. This agent is responsible for the execution of the FCA processes. Because an automated FCA system offers new possibilities for data processing, the FCA process elements are structured different in the MAS than in the manual integration. Although the self-analyzing functionality is added, only four process elements are left in the MAS:
• FCA1.1: Filter out the data of the ERP system
• FCA1.2: Material accounting
• FCA1.3: System accounting
• FCA1.4: Generate and analyze reports The result:
The result of this research is a design that can be seen as a framework for a possible agent- based approach to use FCA in the SCOR-model. It is a simplified model but it covers the heart of the matter, which makes it useful for the next step in the Supply-Chain Flow Cost Accounting Process Program.
Table of contents
Management summary...2
Table of contents ...4
Chapter 1: Introduction to the research ...6
1.1 Singapore ...6
1.2 SIMTech ...6
1.3 Description of the research ...7
1.3.1 Material Flow Cost Accounting ... 7
1.3.2 The SCOR-model ... 8
1.3.3 Supply-chain Flow Cost Accounting Process Program ... 8
1.4 Problem definition...9
1.5 Approach to the research ...9
1.5.1 Integration of FCA in the SCOR-model ... 10
1.5.2 Agent technology ... 11
1.5.3 Design of the MAS... 11
1.6 Framework of the research... 11
Chapter 2: Getting familiar with the SCOR-model ...13
2.1 Structure... 13
2.1.1 Processes... 13
2.1.2 Performance measures ... 14
2.2 SCOR-levels ... 15
2.3 Difference with classic process decomposition models... 16
Chapter 3: Getting familiar with the FCA method ...17
3.1 Flow oriented costs... 17
3.2 The basic idea of FCA ... 18
3.3 Method of Flow Cost Accounting ... 19
3.3.1 The information base ... 19
3.3.2 Accounting elements in flow cost accounting... 21
3.3.2.1 Material flow accounting... 21
3.3.2.2 System cost accounting... 23
3.3.3 Results and report forms in flow cost accounting... 25
3.3.3.1 Material flow model with data... 25
3.3.3.2 Flow cost matrix ... 25
3.3.3.3 Flow cost report... 26
Chapter 4: Integrating FCA in the SCOR-model...27
4.1 The information base... 27
4.1.1 The database in SCOR... 27
4.1.2 The material flow model in SCOR ... 27
4.2 The accounting elements ... 30
4.2.1 Material accounting in SCOR ... 30
4.2.1.1 Data-cards... 31
4.2.1.2 Material transformation during the process ... 32
4.2.1.3 Material flow model with data... 33
4.2.1.4 Material costs ... 33
4.2.2 System accounting in SCOR... 34
4.2.2.1 System cost delimiting ... 35
4.2.2.2 Allocation of system costs... 35
4.2.2.3 System cost apportionment ... 35
4.3 Results and report forms ... 36
4.3.1 Structure of the data-cards. ... 36
4.3.2 Material flow model with data ... 37
4.3.3 Flow cost matrix... 37
4.3.4 The flow cost reports ... 38
4.4 FCA process elements ... 38
Chapter 5: Getting familiar with agent technology...40
5.1 Characteristics of an agent ... 41
5.1.1 Difference between agents and objects ... 41
5.1.2 Agent Building Shell ... 42
5.2 Using agents in the supply-chain... 42
5.2.1 Infrastructure of a MAS ... 43
5.2.2 Design issues for a multi-agent supply-chain system ... 43
5.2.3 Characteristics of a multi-agent supply-chain system... 44
Chapter 6: Designing a MAS for the FCA method ...45
6.1 Constructing the SCOR-foundation ... 46
6.1.1 The SCOR-agents... 46
6.1.1.1 Allocating the SCOR-processes to the agents ... 47
6.1.1.2 Description of the agents... 48
6.1.2 Coordination in the system ... 51
6.2 Designing the FCA module ... 53
6.2.1 Periodic versus dynamic ... 53
6.2.2 Allocating the FCA-process to agents ... 54
6.2.3 Process elements in the MAS... 54
6.2.4 Functions of the FCA-agent ... 54
6.2.4.1 FCA1.1: Filter data out of the ERP system ... 55
6.2.4.2 FCA1.2: The Material accounting ... 56
6.2.4.3 FCA1.3: System accounting... 57
6.2.4.4 FCA1.4: Generate and analyze reports... 58
Chapter 7: Concluding remarks ...62
Chapter 8: Reflection to the research and recommendations ...64
8.1 Achievements... 64
8.2 Reflection on the research... 64
8.3 Validity of the design ... 64
8.4 What are the next steps? ... 65
Acronyms and abbreviations: ...67
List of references ...68
Chapter 1: Introduction to the research
In this first chapter the motivation for this research will be explained and the environment in which the research is done will be described. In paragraph 1.1 Singapore will be introduced, paragraph 1.2 is about SIMTech. Next the motivation for the research will be described in paragraph 1.3 and finally, in paragraph 1.4, the framework for the research will be set up.
1.1 Singapore
In 1965 Singapore obtained independence form Malaysia after dissolution from the British Empire. The initial outlook was unpromising, since the island is very small, only 580 square kilometers, and apart from a good harbor it has no natural resources. Despite this the Singapore government managed to build a strong, export-led manufacturing and service economy. Today Singapore is one of the world’s busiest ports.
Since its independence Singapore has turned into a fast-paced modern city, which is one of the most accessible in Southeast Asia. The multicultural population numbers 4.3 million and comprises the main ethnic groups of: Chinese (76.6%), Malay (14%), Indian (7.9%) and the rest (1.4%). Buddhists, Muslims, Christians, Hindu’s, Sikhs, Taoists and Confucianists live and work together on this small island. The official languages besides English are Chinese, Malay and Tamil. In every day life they blend together to form the typical Singlish.
Singapore has a highly developed and successful free-market based economy, an open and corruption-free environment, stable prices and the fifth highest GDP in the world. Exports and services are the main drivers of the economy. The country is positioning itself as the region’s financial and high-tech hub. In this setting, Gintic Institute of Manufacturing Technology, which has been renamed to SIMTech in 2002, was formed in 1993.
1.2 SIMTech
Singapore Institute of Manufacturing Technology (SIMTech) is Singapore's national applied research and development (R & D) institute in manufacturing technology funded by the Agency for Science, Technology and Research (A*STAR), previously National Science &
Technology Board. It was formed in 1993 upon the merger of the Singapore Institute of Computer Integrated Manufacturing and the Institute of Manufacturing Technology.
Since then, SIMTech has completed about 600 projects for almost 400 companies, which include both multinational corporations and local enterprises. Some of these collaborations have resulted in technological advancements and breakthroughs as well as innovative solutions to industrial problems. The institute provides training programs too: This includes cooperative programs, technology infusion courses, seminars and workshops in specialized fields.
In the progressive and modern country of Singapore, education is very important to keep up with the latest technology. Besides a lot of production- and manufacturing companies, Singapore has two universities: The National University of Singapore (NUS) and the Nanyang Technological University (NTU). SIMTech is located at the campus of the NTU.
Although it was originally setup as a research institute within the NTU, now it is an independent institute. But still there are students from both NUS and NTU attached to SIMTech.
Besides the local universities, SIMTech has contacts with international renowned universities and institutes in Switzerland (University of Neuchâtel), Australia (Cooperative Research Center for Alloy and Solidification Technology), Korea (Korean Advanced Institute of Science and Technology), Japan (National Industrial Research Institute Nagoya) and with
China (Hua Chong University). Contacts with the United States are well established with Stanford University, Cornell University and with the Georgia Institute of Technology.
SIMTech works with these organizations through collaborative research projects and staff exchanges. The results of SIMTech’s R&D activities are published at international conferences and in prestigious journals, patented and published as SIMTech Technical reports.
To contribute to the research manpower resource for Singapore, SIMTech sponsors full-time Masters and PhD students in the universities to work on topic of relevance to the manufacturing industry.
SIMTech’s major tasks are to perform applied Research and Development in processes and technologies to propel Singapore’s manufacturing industry into the 21st century. It also aims at upgrading the local-based manufacturing companies to remain competitive, and at transferring technical know-how from applied R&D to the local-based manufacturing community to bridge the technology gap between small enterprises and multinational corporations. The Technology Infusion Program (TIP) is a training program, which organizes courses throughout the year and aims at creating a pool of manpower to transfer knowledge and skills to the industry.
SIMTech sees that it has an active role to play to sustain manufacturing’s contribution of 25%
of Singapore’s GDP in the future. As Singapore’s industry undergoes structural changes, SIMTech’s role and activities become more important. To upgrade industries for the future, the industry identifies emerging technologies and customizes these for applications in the industry. SIMTech is engaged in improving manufacturing technology of the whole spectrum of Singapore’s industries. Or as the institute’s mission says:
‘SIMTech’s mission is to enhance the competitiveness of Singapore’s industries through the generation and application of advanced manufacturing.’
1.3 Description of the research
To compete with the low costs countries in the region, Singapore’s industry has to reduce its production costs. To achieve this, the SCM Center started the ‘Supply-chain Flow Cost Accounting Process Program’, which will focus on the material costs. Although material costs often account for the greatest ‘cost-block’ in manufacturing companies (especially in mass- production), the existing cost accounting methods are frequently not able to display detailed data about the cost structure of the material flow. The processes have lots of waste material and just loose material during the process. A small percentage of savings in this area can be a huge amount in dollars.
To obtain more detailed information about the material costs, a more detailed material flow is needed. A transparent material flow will enable companies to realize potentials and evaluate appropriate measures for improvement. As a result of the more detailed material flow, reducing the material flow quantity for savings will be possible as well. This means that the process costs can be reduced too, because a part of the production capacity is constant used for processing material, what will be wasted later in the process.
The Material Flow Cost Accounting approach makes the material flow more transparent, so the cost structure of the material flow can be fine-tuned and the waste becomes visible.
1.3.1 Flow Cost Accounting
Flow Cost Accounting (FCA) is an essential instrument in a new management approach know as Flow Management (Strobel, M. 2001). FCA was developed in Germany by the IMU institute (Institute fur Management und Umwelt, founded by B. Wagner and M. Strobel) in cooperation with the university of Augsburg. It shows the losses, inefficiencies and opportunities in the material flow, which other traditional accounting methods will not show.
Existing traditional cost accounting methods are not able to provide sufficiently precise data concerning the cost of materials as well as material losses. FCA eliminates this by closely linking quantitative data to the material flows. In chapter 3 this method will be further explained.
1.3.2 The SCOR-model
To understand the objective of this research, it is necessary to understand the SCOR-model.
The model will be shortly introduced here, but in chapter 2 the SCOR-model will be further explained
All the tools the SCM Center develops are in line with the SCOR-model. The SCOR-model (Supply-chain Operations Reference model) has been developed by the Supply-Chain Council, an independent non-profit organization. The members of the council, which are mostly manufacturing companies and research institutes, are maintaining the model and developing it further.
The SCOR-model is a top-down approach for almost every supply-chain. The scope of the model is the supply-chain from the supplier’s supplier till the customer’s customer. The model decomposes the supply-chain by categorizing all the activities in processes, which is coded by strict rules.
The model consists of at least three levels. For all of these three levels performance measures are defined. By decomposing these metrics from the top-level to lower levels, lower level goals can be derived, and the interdependency between the processes becomes clear. Level four and lower are system specific levels, which organizations have to define for their own system.
The first level, the strategic level, structures the supply-chain’s basic processes. Level two, the material flow level, will decompose the level-one processes, and the material flow becomes visible. From this level and below, FCA might be useful.
1.3.3 Supply-chain Flow Cost Accounting Process Program
As stated before, the SCMC has started the Supply-chain Flow Cost Accounting Process Program. Basically, this research is the first step in the program. The objectives of the Supply-chain Flow Cost Accounting Process Program are:
• To further develop the FCA-approach, and in particular the FCA methodology.
• To reduce the quantities of material used.
• From the experienced gained, the intension is then to derive a transferable concept for standardized application and software integration for manufacturing companies. The SCMC wants to develop intelligent software for this application using agent technology.
• Development of concepts for an effective support of these instruments.
• To set up an international network to cooperate and discuss with, to benefit from the possible synergistic effects.
Now the method of Flow Cost Accounting, the SCOR-model and the Supply-chain Flow Cost Accounting Process Program have been briefly introduced, the goal of the project can be defined. This goal is in fact a sub-objective from the Supply-chain Flow Cost Accounting Process Program: The goal of the project is to integrate the method of Flow Cost Accounting in the SCOR-model.
1.4 Problem definition
This research is about how to use Flow Cost Accounting in the SCOR-model. Using FCA in the SCOR-model can reduce production costs, because manufacturing industries can make savings with a more detailed material flow. De Leeuw (1996) distinguishes 3 core types of research: Pure scientific research, Management supporting research and Problem solving research. This research can be defined as a problem solving research. Problem solving research should deliver knowledge that can be practical used by the customers. The outcome of this research has to be transformed in a transferable concept of course, but it is in fact the solution to the problem. In this case, the customers are all the Supply-chain Council members who are using the SCOR-model.
For this research, a traditional problem definition as stated by De Leeuw (1996) will be used.
This problem definition consists of a main question, an aim of the research and some research delimitations.
Aim of the research
To describe a possible integration of Flow Cost Accounting techniques in the SCOR- model, and to design an agent-based software structure for this integration.
Main question
In what way can Flow Cost Accounting be integrated in the SCOR-model, and what would the relating software look like?
Sub questions
• What is the principle of the SCOR-model?
• What is the principle of Flow Cost Accounting?
• What is the principle of agent technology?
• How can Flow Cost Accounting be used in the SCOR-model?
• How can agent technology be used in the integration of FCA in the SCOR-model?
Research delimitations
• The research will take 6 months.
• No software will be developed, only the structure of an agent-based FCA framework will be designed.
• The ‘Return’ processes in the SCOR-model are still under development, and these will not be fully covered in the integration.
• From SIMTech there is the desire to base FCA software on Agents Technologies.
1.5 Approach to the research
FCA is a method on its own; it has no bigger framework in which it operates. To give the method more meaning in a company, it should be integrated in a company-wide framework.
The IMU, the founder of FCA, is negotiating with SAP now (the German software company which offers one of the most used ERP (Enterprise Resource Planning) systems) to integrate FCA in SAP. Another option would be to integrate it in the SCOR-model. The SCOR-model focuses on the flows through the supply-chain too, and besides that it is a framework of relationships among standardized management processes. Integrating FCA in SCOR is not only adding functionality to the SCOR-model, it also offers FCA a framework to operate in.
To integrate FCA in the SCOR-model, knowledge of as well the SCOR-model as FCA is required. Although it is a complicated model, it is not hard to get familiar with the SCOR- model. At SIMTech a lot of information about the model is available. SIMTech is a member of the Supply-chain itself, and plays an active role in spreading the SCOR-model in the region. As said before, the SCMC is organizing workshops and conferences about SCOR and has already done many projects on SCOR with many companies. In this report the latest version of SCOR will be used, version 5.0
It is harder to get information about Flow Cost Accounting than getting information about the SCOR-model. Only a few people are working on Flow Cost Accounting. There are no (english) books about it yet, so very little information is available. The first project about FCA was only launched in 2000 and is still not finished. Basically, getting familiar with FCA techniques is a little research itself.
As said before, no not-German books are available yet about Flow Cost Accounting. So the main information source is the internet. Via the internet a number of organizations and individuals have been approached, and like this much valuable information could be collected. One document in particular seemed very useful, and many parts of the explanation of FCA are more of less copied from that document (Flow Cost Accounting, February 2001, Strobel and Redmann).
It is useful for the SCMC to stay in touch with these FCA experts to build an international FCA network. This is important because the SCMC needs partners for its Supply-chain Flow Cost Accounting Process Program.
1.5.1 Integration of FCA in the SCOR-model
When the model and the method are known, they have to be connected to each other. It is important to keep in mind that the SCOR-model is the dominant factor in creating this connection. FCA techniques must be integrated in the SCOR-model and not the other way around. After all, the SCOR-model is the framework the SCMC is working with. A second condition is that the SCOR-model itself does not change. Members of the Supply-chain Council must have the choice to use or not to use the opportunity of flow cost accounting when they use SCOR in their organization. So the integration of FCA must have a modular character. Finally, it is not necessary to copy the method of FCA exactly as the IMU has developed it: there is the freedom to adjust the method to make it fit better in the SCOR- model.
The opposite is true for the SCOR-model: The model itself should remain unchanged. To integrate FCA techniques in the SCOR-model the FCA method must fit in the structure of the SCOR-model. This means there are some delimitations. For instance, the SCOR structure or adjust the SCOR language cannot be changed. The FCA module should fit seamless in the language and structure of the SCOR-model.
For integrating FCA in SCOR, the FCA-method will be decomposed and all the separate parts will be linked, converted or aligned with the SCOR-model. Very important are the data- sources of FCA. To apply the method to the SCOR-model, the required data have to be available in the SCOR-model. I will design some data-cards and cost-tables, which will guide the user through the process of Flow cost accounting. The result should be a manual integration of FCA in the SCOR-model.
1.5.2 Agent technology
When the integration of FCA in the SCOR-model has been described, the latitude of the method becomes clear. FCA uses a huge database, and filling all the data-cards and cost tables manually for each process will take a lot effort. So, when an organization will use FCA in practice, the whole technique has to be computerized. A huge amount of calculations must be made, and it is obvious that an IT-system is much more efficient for this purpose. But what should this system look like? Many configurations and architectures are possible. However, SIMTech has intimated the wish to use an agent-structure for this particular system.
Agent technology is a rapid evolving technology, which is said to have a big future. The self- adapting systems could do a lot of work that is now done by humans. All agents have their own goal, and to achieve the common goal of the company they have to negotiate with each other. The system is not prescribing exactly how to handle in particular situation, but provides the agents with the capacity to judge the situation, and then act in a way that seems to be best.
The new system will automate not only FCA, but also the SCOR-model. Intelligent agents will take over the scheduling and rescheduling of the plans.
Besides the area of FCA, agent technology could be used for many more supply-chain-related applications. Before I will use this technology, I will explain the basics of agent technologies.
This explanation comprises in general what agents are, and will focus on Multi-agent systems and how to use them in a supply-chain environment. Note that this will be just a brief introduction into agents to explore the opportunities for developing agent-based FCA software. For further explanation of agents see the literature I refer to.
1.5.3 Design of the MAS
After this introduction into agent technology, the principles of agents and multi-agent systems (MAS) should be clear. When we know what agents are, and we are aware of the difficulties of designing a MAS, I will start to construct a structure for an agent-based integration of FCA in the SCOR-model. In this design I will describe a possible structure of such a MAS, define the different agents and discuss their responsibilities and tasks.
1.6 Framework of the research
In the approach to the research it is easy to recognize the various stages of the research. First knowledge of the SCOR-model as well as FCA is required. So, the first step will be to get familiar with both techniques: Chapter 2 will explain the SCOR-model, chapter 3 will explain the FCA method.
The next step is to describe the integration of the two models, what will be done in chapter 4.
The third step, in chapter 5, will be to explore the area of agent technologies. After that, in chapter 6, an agent-based framework will be designed to use FCA in the SCOR-model. In chapter 7 I will make some concluding remarks on the research, and finally chapter 8 will contain a reflection on the research and some recommendations for the further use of this result in the Supply-chain Flow Cost Accounting Process Program.
A representation of the framework of the research is shown in figure 1.1 on the next page:
Figure 1.1: Structure of the research Getting familiar with the SCOR-model
Getting familiar with FCA
Integrating FCA in the SCOR-model
Getting familiar with Agent Technology
Designing a MAS for the FCA method
Concluding remarks and recommendations
Reflection on the research and recommendations
Chapter 2: Getting familiar with the SCOR-model
The SCOR-model has been developed to describe the business activities associated with all phases of satisfying the customer. The SCOR-model is a framework for supply-chain management. The model itself contains several sections and is organized around the five primary processes (process types):
• Plan
• Source
• Make
• Deliver
• Return
Figure 2.1 shows these processes:
By making use of these building blocks, any organization can be described using a common set of definitions. This way, disparate industries can be linked to describe any supply-chain.
This method is especially useful for big manufacturing companies, but theoretically every organization can be described with the model. The model spans all customer interactions, all physical transactions and all market interactions. The model describes only direct supply- chain-related processes, so for example marketing, human resource, R&D activities and some elements of post delivery are excluded in the model.
The model is designed to be applicable in many different industries and does not attempt to prescribe how organizations must conduct its business. The model defines three process levels, and gives organizations the freedom to extend the model to lower levels. But every organization that uses the model should at least extend the model to level four, using organization specific processes, systems and practice.
2.1 Structure
2.1.1 Processes
Besides the five process types as listed above, it is useful to distinguish three different types of processes: Planning -, execution- and enable-processes. A planning-process (the first of the
Return Return
Deliver Make
Source
Return Return
Deliver Make
Source
Return Return
Deliver Make
Source
Return Source Return
Deliver
Plan
Plan
Plan Plan
Plan
Company Customer
(Internal or External) Supplier
(Internal or External)
Customers customer Suppliers
supplier
Figure 2.1
five management processes) is a process that balances the available aggregated resources with the aggregated required demands. Execution-processes are triggered by planned or actual demands, and change the state of products. Enable-processes, finally, create the environment in which the other processes take place. Enable-processes prepare, maintain and manage information and relationships upon which planning and execution processes rely. In fact each of the five management processes includes enabling-processes.
In the model a set of standard notation is used. All process types, categories and elements have their own logical code. So plan processes are indicated by ‘P’, all source processes by
‘S’ etc. There are two kinds of Return processes: DR (Deliver Return) and SR (Source Return). Enabling processes are indicated by ‘E’, followed by the code for the planning or execution process. So for example ‘EM’ indicates an enabling process that is associated with a Make-process. At a lower level for example an element coded like P1.3 could be found. P indicates a planning element, the ‘1’ is a level-two code, which means (in case of a planning element) ‘plan supply-chain’. Finally the ‘3’ means ‘balance supply-chain resources with supply-chain requirements’. The SCOR model maps all these different elements with their interdependencies, required inputs and outputs in a structured way. Figure 2.2 shows the relation between process types and process categories. Each of the categories consists of process elements.
2.1.2 Performance measures
The SCOR-model has defined almost 150 performance measures divided over three levels. It is important to note, that like the processes, these metrics are intended to be hierarchical as well. Level-one metrics can be decomposed in level-two metrics, and level-two metrics can be decomposed in level-three metrics. The top management defines the business objectives and transfers them into target scores for the top-level metrics. By decomposing the top-level metrics in multiple lower-level metrics the lower-level targets can be derived. In discussing the targets of the lower-level metrics, the interdependency between the different metrics becomes clear. The calculations of the metrics are very strict defined. This way, organizations can be compared in benchmark-studies (like SIMTech does) and best practices can be identified. Best practice is the co-operative way in which firms and their employees undertake business activities in all key processes. These practices, when effectively linked together, can be expected to lead to sustainable world-class outcomes (Australian Manufacturing Council, 1994).
The metrics measure the areas as shown in figure 2.3, divided into customer-facing attributes and internal facing attributes. Each of the metrics will measure only one of the attributes. The
‘Return’ processes are quite new in the model, and no performance measures for these processes are developed yet.
Process types
Plan Source Make Deliver Return
Process categories
Enable Execution
Planning P1 P2 P3 P4 P5
////////// S1-S3 M1-M3 D1-D3 DR, SR
EP ES EM ED ER
Figure 2.2: The relation between process types and process categories
2.2 SCOR-levels
As said before, the SCOR-model consists of three levels. Level one contains the scope for the SCOR-model. Here the basis of competition targets are set. An organization must identify which of the five processes are represented in their supply-chain configuration. Not so many configurations for a single organization are possible: Almost every organization has Plan -, Source -, and Deliver processes. Only a couple of configurations are left then. A pure trade company for instance has no ‘make’ processes, and maybe even no ‘return’ processes. This may seem very easy to distinguish, but for big organizations this can be very complicated. A big manufacturing company can have many production sites, consisting of more organizations. To distinguish and connect all the process types of all the organizations within the scope of the model can be a very hard process. In practice, this model does not fit on a single piece of paper; sometimes an entire wall is not enough.
Level one has 13 metrics, which are typical assigned to the planning of the whole supply- chain (process category P1).
At level two the processes of level one are further categorized. The SCOR-model defines twenty-five categories of core processes. For instance all the execution processes can be distinguished between ‘make-to-stock’, ‘make-to-order’ and ‘engineer-to-order’. The plan and enabling processes have different categories as well, and this way the supply-chain can be
‘configured-to-order’ (see figure 2.1). In this stage the material flow will become visible.
At level three the process categories of level two are decomposed. Each of the twenty-five categories is divided in sub-processes. For each of these process elements the relation between the inputs and outputs of this process element on one side, and other process elements on the other side are described. In fact, the organization’s ability to compete successful in its chosen market is defined. This consists of:
• Process element definitions
• Process element information
• Inputs and outputs
• Process performance metrics
• Best practices, where applicable
• System capabilities required to support best practices
• Systems and tools
From level four, the supply-chain levels are system specific. At level four, the process elements from level three are decomposed in system specific process elements. Organizations implement specific supply-chain management practices at this level. The systems and process-specific details can be defined at this level and lower. Figure 2.4 on the next page shows the three levels of the model plus the first system specific level. The element names at level three and the decompositions are just examples.
Internal-facing Performance
attribute Reliability Cost Assets
Customer-facing
Metrics
Responsiveness Flexibility
Figure 2.3: Five categories for the performances measures
3
2.3 Difference with classic process decomposition models
The SCOR-model decomposes the supply-chain into processes. Breaking down difficult business functions and processes into simpler activities is known as process decomposition.
The SCOR-model uses this technique, but in a different way than the traditional process decomposition. Traditional process decomposition looks at a process with a decreasing aggregation level to understand better what is happening in a particular process, and process decomposition models are developed to represent one specific configuration of process elements. However, the SCOR-model is a process reference model. The SCOR-model is prescribing out of which processes an entity in a supply-chain should exist, and is reconfigurable instead. The three process levels of the SCOR-model are connected through the SCOR process-language and performance measures, as well horizontally as vertically.
This makes the SCOR-model not a simple variant to the traditional decomposition model, but a process reference model that provides a language for communicating among supply-chain partners.
Top-level (Process types)
Plan
Source Make Deliver
Return Return
Level 1
Process element level (Decompose
processes)
Implementation level (Decompose process
elements)
P1.1 P1.2
P1.3 P1.4
Level 4 Level 3
Configuration level (Process categories)
Level 2
Level Schematic
SCOR
Figure 2.4: The hierarchical structure of the SCOR-model
Chapter 3: Getting familiar with the FCA method
At this moment a lot of effort is put in developing environmental accounting methods, Flow cost accounting is one of them. Strobel and Redmann (2001) give the following definition:
‘Flow Cost Accounting cuts costs and relieves stress on the environment by means of an accounting approach based on the actual flow of materials.’
Flow Cost Accounting is not only targeting on relieving the environment, but especially on reducing production costs. This makes the method very attractive to the SCMC because the interest of Singapore’s industries is especially in reducing production costs, relieving the environment is only second.
3.1 Flow oriented costs
In flow-oriented costing there are 6 cost segments to be dealt with (Krcmar and others, 2000).
The material costs are divided in two segments, namely those costs for materials which physically enter the product (including packaging), which is cost segment 1, and those for materials which are physically contained in material losses: Segment 2. All the costs that incur in the course of in-house handling of the material flows are called system costs (e.g.
personnel costs, depreciation). These costs are divided in three segments. The first, segment 3, contains those system costs incurred in manufacturing the product. Segment 4 contains those system costs used in handling material losses before these losses incur. Segment 5 consists of the system costs that incur in handling material losses. The last group, segment 6, contains the disposal costs. Figure 3.1 shows these six segments. So far Flow Cost Accounting is the only method to takes all six costs segments in account.
6. Disposal costs
5. System costs after material losses occur
System costs Material
costs
3. System costs for products
1. Material costs in the product
2. Material costs in material losses 4. System cost before material losses occur
5
6
Figure 3.1: Flow oriented costs
3.2 The basic idea of FCA
The instrument of flow cost accounting (see also Wagner, B. / Strobel 1999, Hessisches Ministerium für Wirtschaft 1999, Strobel 2001) shifts a company’s in-house material flows into the center of the costs analysis and attempts to make these flows transparent beginning- to-end in terms of their effects on costs, in full and flow-specific. This transparency can make a meaningful contribution to clarifying the complex relationships of effects operating within the material flow system and thus create a comprehensive database for realizing potentials and evaluating appropriate measures for improvement. Flow cost accounting may reveal that a measure designed to raise efficiency on a production system leads not only to lower costs in the actual materials but also to lower costs in materials handling and waste disposal. In order to reveal potentials and comprehensively assess the cost effects of planned measures, first of all the effect of each measure on the whole material flow system must be calculated and evaluated.
In Flow Cost Accounting the values and costs of material flows are divided into the following categories:
1. Material 2. System
3. Delivery and disposal
In the material flow, energy is included as well. Especially since energy is often in material form when it enters the company (e.g. coal, oil, gas). So from now on when is spoken of material, energy is included. In the next following sections the three different flows will be discussed:
1. Material
For the purpose of calculating the material values and costs, detailed knowledge of the physical quantities of materials involved in the various flows and inventory is needed.
Usually, the existing planning systems provide a database with such information. This database needs to be adjusted and expanded. Otherwise the database must be created. Based on this differentiated flow quantities, one can proceed to make valuations in terms of prices.
In this way, the material value of the different flows can be determined. The last step to determine the material costs is to define which material flows are cost-relevant. The transparency of knowing values and costs purely for materials already creates, at acceptable expenses and effort, new access to the largest costs block.
2. System
Once again: All the costs that incur in the course of in-house handling of the material flows, including labour, are called system costs. For the purpose of assigning the system value and costs, material movements have to be treated as cost drivers. Each in-house material flow can be seen as a cost carrier for apportioning system costs according to cause. System costs are accounted to the outgoing material flows and then passed on as system values to the subsequent flows and inventories.
3. Delivery and disposal
When flows leave the company, delivery and disposal costs must be assigned. This category includes all costs incurred in ensuring that material leaves the company, i.e. not only transport costs for products but also, in particular, the external costs for disposing of waste and the fees for waste water and effluent.
In traditional cost accounting, material costs and system costs are mixed together after the first processing. So it becomes impossible to list costs and values separately according to the three afore-mentioned categories, material movements and inventories. The result of Flow
Cost Accounting is the beginning-to-end transparency showing quantities, values and costs of material flows, separated into the three categories ‘material’, ‘system’ and ‘delivery/disposal’.
Figure 3.2 shows the basic idea of Flow Cost Accounting:
3.3 Method of Flow Cost Accounting
The method of Flow Cost Accounting consists of a couple of components, see figure 3.3 on the next page. The data flow and the sequence of flow cost accounting run from the stipulation of the necessary database by way of the accounting elements right through to the various results and report forms. However, in the opposite direction, in response to specific demands made of the results and report forms, there may also be adaptations regarding the accounting elements or even the database specified. In the following sections the component parts of flow cost accounting are described in detail.
3.3.1 The information base
The information base consists of two elements, which are the material flow model and the database. The material flow model maps the structure of the material flow system, and the database contains the data needed to quantify the material flow model. The database refers
Figure 3.2: The basic idea of Flow Cost Accounting Supplier
Disposal Customer
Company
Raw materials store
Intermediate product store
Production Quality
control
Waste disposal system
Outgoing goods store
Material
Delivery/
Disposal
System Source: Strobel/ Redmann, 2001
both to material flows as inventories, and also includes other relevant system data. It is used as basis for calculating the quantities, values, and costs assignable to the flow model.
1. Material flow model
For the purposes of describing a company in terms of a material flow system, we use the material flow model. Using this model it is possible to map out all the works-specific material flow structures in a comprehensive and beginning-to-end manner. This helps to raise the level of transparency of all in-house inter-connections and relationships. It is quite similar to figure 3.2, but in the material flow model the flows are not yet separated in the different categories.
In setting up the model the first step is to define the system boundaries of the material flow system and thus of the flow model. The system boundaries should enclose all those company units within which material flow transparency is to be enhanced. The other basic representation elements in a material flow model are, besides the system boundaries, the quantity centers and the material flows (cf. Krcmar et al 2000). Figure 3.2 has showed the material flow model already.
Quantity centers
Quantity centers are spatial and functional units at which material is transformed physically or exists through time (e.g. stored, processed, tested, or sorted). Typical examples for in-house quantity centers are a company’s storage areas, production areas, filter systems, etc. External quantity centers (e.g. supplier, customer, final disposal agent) form the flow-relevant outside world for the company, i.e. the source and destination points of the company’s input and output flows when these pass beyond the system boundaries. External quantity centers should therefore always be incorporated in the material flow model.
Material flows
Material flows represent the structure of possible material movements from one quantity center to any other. For the purposes of flow cost accounting it is sufficient, if just one material flow is entered in the model between the two quantity centers in question without this having to be further differentiated. To ensure that quantity centers and material flows are
Results and report forms Flow model
with data
Flow cost report Flow cost
matrix
Accounting elements
Material flow accounting System cost accounting
Information base
Material flow model Database (e.g. SAP R/3, Oracle or Baan)
Figure 3.3: The components of Flow Cost Accounting
clearly and unequivocally identifiable, these should each be assigned a code number. Quantity centers and material flows should be given these code numbers in the model, thus giving the data in material flow accounting a direct reference to the flow model.
As well as using graphical representation in a flow model, the results of the material flow mapping are also documented in table form. These ‘material flow tables’ and ‘quantity center tables’ may also contain additional relevant information (e.g. areas of responsibility, competences, computer systems, etc.).
‘The primary objective of flow modeling is to represent the essential material flow structure in a company by aggregating in some suitable form all its in-house quantity centers and material flows.’
2. Database
A comprehensive database is similarly a prerequisite for material flow accounting. The quality of the database is decisive if flow cost accounting is to be meaningful at all.
Database for material flow accounting
The database required for material flow accounting must contain not only prices and the net bill of materials (BOM) but also, an essential component, the quantity data. This quantity data should always refer to a period. Quantity data refer both to material movements and to material inventories. Access to existing quantity data is not usually a problem. Material management and production planning and control systems often contain the majority of the quantity data needed. The quantity data describing material movements can be used to allocate the appropriate quantity data to material flows in the material flow model.
The quantity data describing material inventories can be allocated as start and end inventories to the quantity centers in the material flow model. The standard ERP systems refer to inventory quantities per storage location. It is useful therefore if storage locations and quantity centers are somehow coordinated. Inventory data is recorded per storage location as a function of stock-taking, either in a complete count or on a sampling basis.
Database for system cost accounting
The database required for system cost accounting comprises system costs and system data.
The database records for system costs comprise information regarding the type of costs, e.g.
personnel costs or write-downs, and allocation options revealing where the costs were incurred. The system data provides quantitative information regarding certain system characteristics (e.g. surface areas, number of plant units, etc.) and thus forms the basis for apportioning system costs to the material flows according to the appropriate rates.
3.3.2 Accounting elements in flow cost accounting
The accounting elements can initially be divided into material flow accounting and system accounting. Material flow accounting is used to check the database with respect to its consistency and to assign data to the flow model. Material flow accounting comprises the following individual elements: material flow quantity assessment, material flow valuation and material costing. System cost accounting is based on material flow accounting and used in a multi-stage procedure to assign system costs to the material flow model. It contains the following elements: system costing delimiting, system cost allocation and system cost apportionment. Both material cost accounting and system cost accounting will be explained here.
3.3.2.1 Material flow accounting
Material flow accounting is used to quantify the material flow system on a per-period basis.
Material flow accounting comprises the following inter-linked computing elements (see also
Strobel, M. / Wagner, F. 1999) - material flow quantity assessment, - material flow valuation, and - material flow costing. Material flow accounting consists of three steps:
1. Material flow quantity assessment
Material flow quantity assessment is the fundamental data structure of material flow accounting. Its first function is, for each material flow in the material flow model, to allocate per individual movement per material (i.e. per material number, article number, part code number, etc.) the quantity that has actually flowed per period in physical units (e.g. pieces, kilograms, meters, etc.). This quantity data may be acquired as part of some company activity, or transferred from the company’s existing database, or subsequently calculated in material flow quantity assessment.
Its second function is to allocate per quantity center the start and end inventories for the period concerned.
The special feature of material flow quantity assessment is that it views material not only per processing stage (i.e. as intermediate product, or finished product, or unsorted waste) but, parallel to this view, shows all materials in terms of incoming materials (incoming materials are all those materials that enter the company from an outside source). Incoming materials are tracked from this point on, through all the material flows and all the quantity centers, right up till when they finally leave the material flow system. Here, and similarly in the beginning-to- end recording of materials used, materials can only be aggregated if they are recorded in the same units of measurement.
Based on these incoming materials, it is now possible to make a difference calculation per period in line with the material flow structure of the material flow model (according to the laws of thermo-dynamics, material and energy may be transformed but never generated of destroyed). It is important to note that this should be done for each type of material or semi- product separately to achieve the goal of a more transparent material flow. This difference calculation shows the difference per quantity center and per material used between the start and end inventories and between all arrivals / departures per quantity center. Assuming postings at the quantity centers are perfectly exact, this difference calculation should produce the result ‘zero’. In practice however, the result is often far from zero. See figure 3.5 on the next page for an example (the result is not zero).
Such deviations point either to inconsistencies in the database or to material movements which the database does not contain. Such missing material movements are usually material losses at quantity centers that have to be allocated to individual material flows in the material flow model.
Raw materials
store
S.I.
5
E.I.
18
Material losses
220 205
2
Figure 3.5: The difference calculation
S.I. = Start inventory E.I. = End inventory
