IEEE TRANSACTIONS ON ENGINERING MANAGEMENT, VOL. 43, NO. 4, NOVEMBER 1996
418
The Value Chain Component in a Decision Support System: A Case Example BjQm Haavengen, Dag H. Olsen, and James A. Sena
Abstract- This paper discusses the contribution and use of the value chain component within a prototype decision support system (DSS). Here we focus on the inclusion of Porter’s value chain within the general design of the DSS for a purchasing manager in the Norwegian wood processing industry. We will discuss the value chain and cost driver parameters as well as those design principles that we found significant for the creation, inclusion, visualization, and use of the value system. We will summarize related research that we conducted concerning the analysis and diagnosis of decision processes for this particular decision maker and his organization and the formulation of a prescriptive model. We then will discuss some of the general design and implementation features of our prototype DSS. The goal of this DSS is to present a tool for the decision maker that addresses qualitative decision making and improve the user’s mental model and modeling capacity.
I. INTRODUCTION
M
ANY decision makers in organizations deal with problems that are difficult to define because of high complexity, lack of structure, or vagueness in their formulation [ 11. In many instances, decision making is based on information that is primarily qualitative. To model decisions that are not solely based on computations we need to assess how decision makers behave, not how they ought to behave. In this paper we use a company as a case to portray our design and more specifically the inclusion of a value chain analysis for a DSS prototype that treats qualitative decision making parameters. The manager of a purchasing division in the Norwegian wood processing industry has been selected as a case to illustrate our approach. We employed Cyert and March’s four descriptors [2] to differentiate actual decision making from rational decision making. The four descriptors are local rationality, uncertainty avoidance, local search, and simple learning. They provided a guide for what to search for when diagnosing decision making behavior and represent a basis for rational simplifications on the part of decision makers. Our particular manager has a unique role within his organization. He operates as a strategic manager, a middle manager, and as a quasi-independent contractor. He communicates and Manuscript received July I , 1994; revised August 1994. Review of this manuscript was arranged by Editor E. Geisler. B. Haavengen is a software engineer with Konigsburg Offshore A.S., Konigsburg, Norway (e-mail:
[email protected]). D. H. Olsen is with the Department or Information Systems, Agder College, N4604 Kristiansand, Norway. J. A. Sena is with the College of Business, Mangement Information Systems, California Polytechnic State University, San Luis Obispo, CA 93407 USA. Publisher Item Identifier S 0018-9391(96)08387-0.
works directly with vendors. His customers are the company units themselves. Given this unique positioning the manager needs to explicitly consider the value chain in order to fully support his decision making processes. This decision maker typifies an increasing emphasis on boundary spanners between organizations. The performance of such a decision maker is essential to the efficient and effective operation of the organization [ 3 ] .The term boundary spanning refers to [4]: . . . those employees of the firm whose positions involve facilitating product, service, and information flows within the distribution channel through personal contacts with associated channel members. Much of the administration in complex organizations is the management and exchange across organizational boundaries
~51. A major part of the gain from implementing this DSS we envisioned to be the improvement of the user’s mental model and his modeling capacity wherein he can address hs boundary spanning requirements. The use of the system ought to contribute to the user’s learning about how he should solve his problems in a wide sense. In our prototype design his learning was envisioned to be along the following dimensions: learning to see his problems in a more global organizational view as opposed to local rationality; learning to better master uncertainty as opposed to avoidance; increasing his understanding of appropriate levels for information search and exchange; increasing his learning from and about primary decision problem domains. While the system is designed to support the existing decision process, the use of the system ought to stimulate and channel his behavior toward a prescriptive process. In line with this discussion the decision process needs to be in focus. The goal of a DSS that addresses qualitative decision making ought to contribute to a better decision process and in this way improve the quality of the decisions made. Instead of decision making, we should probably talk about unfolding a decision in the Japenese sense, with emphasis on the process rather than on the act of decision making 161. 11. OUR APPROACH
The decision maker in this study is the manager of the purchasing division in a Norwegian wood processing company. The division is a decentralized unit that functions as a standard cost center. The division supplies other divisions
0018-9391/96$05.00 0 1996 IEEE
HAAVENGEN et al.: THE VALUh CHAIN COMPONENT IN A DECISION SUPPORT SYSTEM
419
Transport of Paper to Market
U plant
Fig. 1. PhyGcal flow of wood.
of the company with roundwood and industrial wood chips. The company itself is the major buyer of pulpwood within the region that the purchasing manager operates. It owns some forest properties and has some logging of its own. Roundwood from its own properties constitutes about 4% of the required roundwood. The roundwood market is characterized by a large number of small suppliers. In the Norwegian supply area there are about 20 000 forest owners. The average size of the suppliers’ properties is SO hectares and the average roundwood harvest during the year is about 100 m3. The logging season is short for the major part of the suppliers-only a few months. This leads to an uneven supply of roundwood during the year. The major part of the forest owners receive a small portion of their income from the supply of roundwood and do not log every year. Their roundwood harvesting quantity depends on the spot prices of lumber and their expectations about future prices over the succeeding periods and years. The minority of the forest owners depend on forestry as their primary income. For these owners the logging quantity is more constant (stable) and the logging operation is distributed throughout the year. Other important parameters for the roundwood market are climatic and topographic characteristics. The roundwood market operates under a significant degree of uncertainty. The manager perceives that the success of the purchasing division is in accordance with the attainment of the organization’s goals. The division’s performance is affected by variations in the activity levels of other divisions. Decisions in the purchasing division also affect the performances of other divisions in the organization. A major part of the manager’s activity is dealing with the external environment. He consistently maintains contact with the suppliers and spends considerable
time coordinating his departments activities with the factories. Fig. 1 presents a diagram of the physical flow of wood. We employed a variety of methods to map the manager’s present decision processes and his information requirements. The methods included modified and extended versions of the critical success factors method (CSF) [7], [SI and the Thinking Aloud Event Protocol method (TAEP) [9]-[ 1 11. We chose these methods because they treated different but complimentary aspects of the decision process. The CSF approach focuses on organizational objectives and factors that any manager would perceive to be critical for the attainment of these objectives. The TAEP approach is descriptive and focuses on the individual decision maker’s handling of complex decision situations. This is a cognitive model that examines the decision process in a holistic fashion-from the time the decision maker receives a signal about an external event until he decides on a course of action. We also conducted seminars with the manager to gather information processing and work environment data. The TAEP method was very helpful in describing the value system. Together these two methods gave us a strong base for describing and understanding the interaction between the manager’s perceptions and his behavior. Fig. 2 summarizes the methods that we employed to map the decision situation. The skills and experience of the interviewers complemented the methods: one of the interviewers had formal education and experience in the wood processing industry. This aided us all in constructing the TAEP scenarios and in understanding and defining the specific requirements related to the wood processing industry and the purchasing manager’s needs. Based on our diagnosis and interpretation of the decision situation maps, we compiled a set of observations about this manager. These observations are framed in terms of
IEEE TRANSACTIONS ON ENCINERING MANAGEMENT, VOL. 43, NO. 4, NOVEMBER 1996
420
~
MANAGER
INTERVIEWER
I
Industry Knowledge
Seminars
Skills
1.7
Goals Factors
El.’ - questions
Mental Models
Processes
I-PIJ
L
R Models
- scenarios
Performance
Models
Modeling Skills
Lm Models
Fig. 2. Methods for mapping decision situations
Wall’s [ 121 propositions about administrative decision making behavior. He appears to have a good set of notions about supply and a weaker set about the market. The manager has a good causal understanding from the supply side, where he appears to be able to incorporate different perspectives. From the supply side he is more of a leader than a middle manager, whereas from the market side he is more of a middle manager. The manager’s reception of signals under crisis situations is strong. As we created our design we tried to incorporate facilities that would assist him in forming an opinion about what is happening and what steps are being taken to prepare for anticipated effects. This manager is aware that uncertainty exists in many of his activities and that his work is influenced by uncertainty. The problem is that this manager does not seem to quantify this uncertainty to any degree. The manager has a good picture of his decision making environment but it is still incomplete. We believe that he needs to have mechanisms available that can identify gaps and deficiencies to better complete this picture. In our design we tried to incorporate features that could fill this gap, or at least approach, the development of these mechanisms. We envisioned the system to be one that contribute to the development of improved decision making. Combining the results obtained in the diagnosis with decision making theory and information from trade publications and research studies we developed a prescriptive formulation for the manager. This process is shown in Fig. 3. We started the formulation phase by performing a physical flow analysis of the basic decision making operations exercised by the purchasing manager and activities tied to his division’s regular operations. The physical flow was used as a foundation to establish the operational decision processes. We perceived this to be important in order to provide an initial fundamental
model for the construction of maps describing the purchasing manager’s decision processes. The impetus to describe the physical flow of wood was based on Porter’s value system [13] that details the cost and activity flow progression. We wanted to describe the activities tied to the flow of wood, the entities performing the activities, their costs, and the linkages among the different activities, both inside and outside the company. We used the description of the physical flow of wood and the company’s Proposed System Specification (an internal company document) to unfold operational decision processes. We then created a formalism, expressed in three dimensions, for grouping the decision making processes, in effect, a cube. The cube is shown later in this paper as part of the main form in Fig. 4. On the primary plane or first dimension of the cube were two core functions, the cost and the flow (of wood) about which all decisions, in one way or another, were concentrated. These are the core functions that affect the manager as he conducts his work and which the performance of the division is measured. We established and placed three central activities on the next dimension: coordination, change management, and budgeting and contracting. Coordination is an activity to which most managers allocate considerable time. We chose the term, change management, to encompass a wide focus for handling crises, handling changes, making contingency plans, and performing long-term planning. Budgeting and contracting is a major activity for this manager, especially in the short and medium term. The third dimension is time. Time is a significant dimension in considering differences between short-term and long-term decision making (e.g., flow-the forest life cycle from regeneration to logging is many decades). We concluded that the cube, with core functions, central activities and time as dimensions
HAAVENGEN et al.: THE VALUE CHAIN COMPONENT IN A DECISION SUPPORT SYSTEM
42 I
;[=I
I -1 1~ I
I
Analysis
Value system considerations
Cyert and March' dimensions
Micro factors E i n t ePr v i e w s H Cognitive maps
Formulation of prescriptive model
CSF diagnosis TAEP diagnosis
Formulation products Decision making cube Decision prosess hierarchies + Information flow diagrams Information flow descriptions Coordination grids Value chain analysis
processes
Fig. 3. The formulation of a prescriptive model.
Fig. 4. Main form.
appeared to be an appropriate visualization of the interaction among the various decision making dimensions. 111. ANALYZING THE VALUECHAIN
Porter [ 131 divides the value system into Suppljer, Firm, Channel, and Buyer value chain components. He describes a generic value chain that consists of different value activities
that are technologically and strategically separated. Value activities, the building blocks by which a firm creates a product valuable to its buyers, are divided into two main types; primary activities and support activities. The primary activities in any industry involve five generic categories: inbound logistics, operations, outbound logistic, marketing and sales, and service. Support activities support the primary activities by providing
422
IEEE TRANSACTIONS ON ENGINERING MANAGEMENT, VOL. 43, NO. 4, NOVEMBER 1996
purchased inputs, technology, human resources, and different firm wide functions. In our example we did not divide into specific primary or support activities but instead described the physical flow of wood beginning with the activity of felling and continuing to the transport of paper to market. A cost analysis describes a firm’s cost position. It includes analyses of individual activities and the relationships between the activities affecting costs. These -analyzes should address broad long-term issues. The steps in the cost analysis include: 1) identifying the value chain; 2) describing each value activity’s cost structure; 3) describing how each value activity’s cost behavior may be affected by linkages and interrelationships with other activities within and outside the firm. Performing a cost analysis begins with defining a firm’s value chain and then assigning operating costs and assets to value activities. The cost behavior of a value activity depends on a number of structural factors that influence costs, which Porter term cost drivers. A value activity can be influenced by more than one cost driver. While one driver may exert the strongest influence on the cost of a value activity, several drivers often interact to determine costs. These drivers may either reinforce or counteract each other. Identifying interaction among cost drivers is a necessary part of determining the cost behavior of a value activity , The value chain of the firm described in this study is embedded in a large stream of activities in the forest value system. There are strong vertical linkages between suppliers’, transporters’, the division’s, and its parent company’s value chains. Hence we have to make an upstream cost analysis to reveal the linkages starting with the suppliers’ value chain that creates and deliver inputs and ending with the downstream parent company’s value activities that influence the purchasing division’s value chain. This is important for two purposes: enhancing global rationality and improving coordination. A coordination of activities across organization lines is often necessary. Frequently the company needs on-time delivery; this can serve to reduce the products lead time, quality decay, and capital costs. Through coordination between the company’s and suppliers’ value chains it is possible to benefit both the company and the suppliers. In this section we have defined each of Porter’s cost drivers and related them accordingly to the purchasing manager’s decision environment: We included this analysis in order to illustrate the complexity of the decision maker’s environment and his need for a value system module. I ) Economies or Diseconomies qf Structure: Economies of scale arise from the ability to perform activities differently and more efficiently at larger volumes. Economy of scale must be distinguished from capacity utilization. Increasing complexity and costs of coordination can lead to dichotomies of scale in value activities as scale increases. The firm’s operations are basically decentralized. The forest property ownership situation is different in Norway than in other countries, such as Sweden and the United States. In contrast with their competitors there are many small suppliers
in the region that increase the need for Coordination. Upstream operational activities are not significant driving forces for the attainment of any economy of scale. These operational aspects are delegated to subordinates. Tactical and strategic planning is instead performed by the purchasing manager. With regard to downstream activities the suppliers have their own associations. A major part of the supplier coordination is assumed by the forest owner associations (there are four owner association subunits in the region-one for each county). A subordinate is responsible and handles the transport coordination while the purchase manager treats contracting and the more long term strategic activities related to supply. Scheduling and daily operation are performed by the associations. There is a different downstream situation with the transporters. Each transporter operates as an independent contractors. The operational logistics are again delegated to a subordinate. It is the tactical and strategic planning where the purchasing manager is involved and can influence economy of scale by such means as setting stock levels and stock capacity (size and placement of terminals) or transporting by ship or rail directly to the mills. 2) Learning and Spillover: The cost of value activity can decline over time due to learning that increases its efficiency. The rate of learning varies widely among value activities because each offers different possibilities for learning improvement. Learning is often the culmination of many small improvements rather than major breakthroughs. Learning tends to vary with the amount of management attention devoted to capturing it. Knowledge of how the supply system works is embedded in the purchasing manager and the forest owner associations. The purchasing manager is an expert in his field. Learning is attained by means of small improvements. Each period, as contract negotiations are completed and throughout the year as revisions are made, the purchasing manager learns from his experiences-in a sense he is backcasting. Small improvements in an iterative sense appropriately describe this manager’s learning approach. Spillovers influence the purchasing manager as well. He monitors decisions and changes in both the Swedish lumber industry and in the southern part of his own company. Interpretations and leaming about what worked for them is incorporated into his knowledge about his operational supply area. 3) Patterns of Capacity Utilization: When a value activity has substantial fixed cost associated with it, the cost of an activity will be affected by capacity utilization. Fixed costs create a penalty for underutilization, and the ratio of fixed to variable cost indicates the sensitivity of a value activity to utilization. Different ways of configuring a value activity will affect the sensitivity to capacity utilization. Production for the division and its parent company has significant fixed costs. With respect to the operation and capacity of the paper machines, there are three production lines. These are essentially running at capacity. The machines are marginally flexible-the ability to expand production is limited. The only available operational alternative is to increase the speed of production which would provide only minor improvements.
HAAVENGEN et al.: THE VALUE CHAIN COMPONENT IN A DECISION SUPPORT SYSTEM
Given this operational constraint increasing the number of suppliers, transporters, or the number and size of terminals would not have a significant impact on production quantity (volumes). The number of transporters and the investment in transport capacity has substantial fixed costs. It is extremely important to utilize the transport capacity throughout the year since the trucks are specialized and have no alternative use. 4 ) Linkage(s): within the value chain are relationships between the way one value activity is performed and the cost of performance of another. The term linkage is used to describe relationships within the value chain. Vertical linkage reflects interdependencies between different value chains (e.g., supplier-firm or firm-channel). The cost behavior of a value activity cannot be understood by examining the activity alone. Linkages create the opportunity to lower the total cost of the linked activity. They provide a potential powerful source of cost advantage because linkages are subtle and require joint optimization of coordination of activities across organizational lines. Linkages are an important factor for the purchasing manager. The ability to coordinate the links will affect the wood quality and thereby the production, paper brightness and the use of bleach. There exists a number of horizontal linkages. The stock keeping condition affects the performance in the debarking process. This performance in debarking influences the pulp production; the use of bleaching powder and energy consumption. The apportion in the felling activity performed by the lumberjacks is significant for the value of the final sawn timber. 5) Interrelationships: interrelationships with other business units within a firm affect cost. Sharing a value activity raises throughput in the activity. Sharing is potentially a way to achieve scale by reducing the learning curve. The company shares terminals among several paper mills. The various mills in a particular area are supplied by specific terminals which are controlled by the purchasing division. The division’s existence is an example of interrelationships-a common purchasing function for all paper mills. With employment of recycled paper the transport facilities are shared with the marketing division, by using the same boats for the transporting of newsprint to the market as transporting of recycled paper from the market to the paper mills. Trade-offs between transporting saw timber and pulpwood have to be assessed given existing capacity limits. 6) Integration: The level of vertical integration in a value activity may influence its cost. Integration can reduce cost in a number of ways by avoiding the cost of using the markets, such as procurement and transportation costs, and instead perform the activities internally within the organization. There is a limited possibility for some form of integration. The company owns some forests-about 4% of the supply. The bulk of the supply comes from the forest owners which means that there is no real integration along the supply side. However, the suppliers are dependent on the company. There is some semi-integration that exists since they have only one main buyer of pulpwood in the region. There is continuous but informal information exchange. Once negotiations and contacts have been completed the flow of wood in various
423
forms commences and continues throughout the contract period. The company can take advantage of vertical integration by transporting wood to their own mills that have the capability to produce pulp wood instead of using independent pulp producing facilities. The use of their own forests can be employed to smooth out and guarantee the roundwood supply. The purchasing manager monitors the supply and demand over the contract period. If there is an increased demand that cannot be met by the current contract he may have to consider other steps to meet the demand such as buying wood from Russia or transfer of wood from the company’s division in southern Norway. 7) Timing: Its role in costing may depend more on the business cycle or market conditions than on timing in absolute terms. Timing can lead to either sustainable cost advantages or a short term cost advantage. There are several aspect of timing that affect the purchasing manager. The market demand for recycled paper has infuenced the requirements for a new supply assortment. Over the past several years there has been an increased demand for paper containing recycled paper materials. The buyers are willing to pay a premium for such paper. This requirement for recycled paper is expected to increase for the next several years. The manager realizes that the company needs to take steps to insure a mix of supply that includes recycled paper in order to keep its market position. Increases and improvements in technology and a willingness of the buyers to accept lower quality paper in consideration to some properties (e.g., strength) has made this alternative feasible. This does not go unnoticed by the suppliers. A substitute assortment affects the contract expectations. 8) Discretionary Policies Independent of Other Lkivers: Discretionary policies reflect a firm’s strategy and1 often involve deliberate trade-offs between cost and differentiation. The supply side of the value system is not a competitive market for the major assortments. The company operates as a high volume, medium quality producer, and all production is custom ordered. To insure goodwill with the forest owners, the company takes and applies marginal quantities of pine that yields a lower quality product. This, in turn, increases the production costs but maintains a good supplier relationship. With this policy quality constraints are relaxed. This, to a large extent, influences supply contracts and thereby logistics and stock levels. 9) Location: The geographic location of a value activity can affect its cost as can its location relative to other value activities. Location for the purchasing manager is related to topographic and climatic characteristics. They both affect the supply and transport of wood and ultimately the cost of wood. Logging distribution throughout the year depends on the location which, in turn, affects the quality of wood. Location with its topographic and climatic characteristics is the main influence for the purchasing manager’s DSS need. They are the major factors for uncertainty in his environment. Climatic and topographic characteristics affect-in addition to the cost of logging and transporting-quality decay throughout the year
424
IEEE TRANSACTIONS ON ENGINERING MANAGEMENT, VOL 43, NO. 4, NOVEMBER 1996
and impact the quality of wood in production and the quality of the finished paper. The purchasing manager attempts to control this quality by transporting lumber from the region to the closest factory. He tries to stimulate logging in favorable areas. The main point for the purchasing manager is that the company has to pay the same price for lumber independent of the delivery place or time and the transport cost is paid by the buyer. 10) Institutional Fuctors: Government regulations, work schedules, financial incentives, unionization, tariffs, and other rules constitute a major cost driver. While institutional factors often remain outside a firm’s control, means may exist to influence them or minimize their impact. There are rules that affect the transport of lumber. For example, in the region there are periods with road restrictions that highly influence the transport of lumber. The forest owners are a de facto union and interface with the purchasing manager as a single entity. The authorities can stimulate logging by providing price incentives to log. They can also affect the transportation through road restriction and road standards. The purchasing manager is trained and educated in forestry. As such, he is an expert in his field. However, the manager has no formal training in decision making or management science. He is both the source and expert for knowledge concerning the value chain cost factors. We tried to design the prototype DSS to channel his behavior by helping him structure his problems with a global view. The system was intended to be a specific system to support this expert judgement. We have included certain aspects, specifically the value chain module, with the aim of allowing this manager to take advantage of the value chain approach. Although by formal training he is not familiar with its use we expect that he will become very proficient in conducting a value chain analysis for his decision making environment. The system provides him a tool for a straightforward value chain analysis. The focus is on the decision as opposed to the function. The value chain approach gives the manager the ability to see the global effects of his decisions. Using the value chain module the manager will be systematic. He can describe the linkages and therefore will have a better picture of existing uncertainties. Every activity that he describes will give pictures of existing uncertainties. As he reviews and establishes linkages he will be triggered to search for new and additional information. In the diagnosis we revealed some characteristics of the purchasing manager regarding the supply side. We found that he had an extensive monitoring and analyzing of the supply side. The manager had limited proactive behavior and his problem solving was event driven. His information processing capabilities were comprehensive and he had a good causal understanding about supply. The CSF’s tied to the supply were external and determined, not controllable by the manager. These findings have been considered and been a major basis for the design and implementation of the prototype DSS. The complexity of qualitative data together with the unpredictability of the supply environment, supply planning and scheduling tasks are too complex to be solved by explicit numerical methods alone. To benefit the manager’s extensive
knowledge, accumulated from years of experience in the indusfry and his comprehensive information processing capability, he needs a tool with simple numerical methods. The DSS is constructed to facilitate the power of the manager’s reasoning imethods based on the development of knowledge structures and heuristics that permit effective and efficient access to the appropriate knowledge in response to the supply environment. During development of the DSS we have included these valuable aspects of the manager’s knowledge. By using the system we expect that the manager will enhance his knowledge base by storing his assumptions.
IV. THEPROTOTYPE DSS AND THE VALUE SYSTEM A primary tenant in our design was a graphical user interface that facilitated visual interactive modeling. In our formulation we revealed that the manager has a need for support in the process of utilizing and storing information. We tried to incorporate into the system a means to promote a value system approach, and make the manager focus on decisions as opposed to functions. We tried to stimulate the manager to store information in a way that directly promotes a value system view by encouraging and directing the manager to identify, and retain data in the system, about the significant cost elements and linkages (to other activities) for each value activity. The purchasing manager operates under a Microsoft (MS) Windows-based micro computer system that is connected to a Novel1 Local Area Network (LAN). Based on his computing environment, the company’s computing plans, and this manager’s computing activity and orientation we concluded that the purchasing manager would best be served through a MS Windows environment with a focus on MS software that would interface and be compatible with Excel spreadsheets. For our software foundation we elected to use Visual Basic. Using Visual Basic we were able to adopt an objectoriented approach that permitted us to present or unfold aspects of the system as they were developed or used. The software enabled us to incorporate common Windows objects or controls into our applications. In the following sections we will describe some of the aspects of the prototype DSS design and implementation. The main screen is the entry point to the purchase manager’s DSS. From this screen the manager can select a specific decision activity, such as the budgeting and contracting, or examine the value system module. He also has the ability to go directly to a specific task and scenario in order to continue or resume work from previous sessions. The main entry screen, entitled Main Form, is shown in Fig. 4. For the main screen we chose to portray, not only the menu bar, but also the decision making cube-the foundation for the purchasing manager’s DSS. We considered the cube to be an application specific way to enter a major decision activity. In addition to the cube we also depicted the pull down selection for the budget and contract activity. This is one of the two major activities fro which we created prototype DSS’s. The other was the change management activity. Both of these prototype DSS’s shared and interfaced to the value system
HAAVENGEN et a1 THE VALUE CHAIN COMPONENT IN A DECISION SUPPORT SYSTEM
425
Fig. 5. The value system form.
module that we created and that is described in subsequent sections of this paper. In keeping with the development of an improved mental model we have incorporated the value system option on the menu. We felt that this could be a way for the manager to view the various entities and activities in terms of linkages and influences. V. VALUESYSTEM MODULE The company’s value chain is embedded within the total stream of activities associated with the forestry value system. The value activities are organizationally, technologically and strategically separated and are an integral part of the value chains for the supply, transport and for the company itself. There are strong linkages among value activities within all of these value chains, and interdependencies exist among different value chains (e.g., supply company or transport company). In the value system module the manager judges the way that different value activities contribute to the total cost for the company. He describes the value activities’ cost behavior and their dependencies on a number of structural factors that influence the cost. This description addresses broad issues and has a long-term viewpoint. In this way the manager
improves his understanding of the activities’ cost behavior, their linkages and the existing and potential sources .for cost improvement. A selection of the value system entry brings the form to the front of the screen layout. This form presents the four value chains (roundwood supply, chips supply, transport, and company). One of the chain’s elements is magnified, showing the value activities’ name. For the other value chains the activities are represented as boxes. Fig. 5 displays the roundwood supply value chain magnified where the activity, “Stock at Depot,” is colored. This activity has been selected and those value activities in the value system that are linked to the “Stock at Depot” activity are highlighted using different colors. The various colors illustrate the activity linkage’s cardinality. The value system form has two modes for presentation of information related to value activities. These are the Value activity mode and Linkages mode. The first option “Value Activity” allows the user to select a value activity in the magnified value chain. When activated the form contains a description of the selected activity. This kind of display is shown in Fig. 6. There the user has selected the value activity “Stock at depot” by clicking on the box representing the value activity in the “Roundwood Supply” chain.
BER I 996
426
Fig. 6. Value system form and value activity form
The form, “Value Activity” as shown in the lower part of Fig. 6, presents a description of the individual activity that the user selected in the “Value System” form. The description contains a general explanation of the value activity’s performance and cost structure. It can also include references to other sources for more detailed information, The “Value Activity” form allows the user to edit and make changes in the description. He is also able to add or remove value activities in all value chains. As new activities are created or existing activities deleted the value chains in the “Value System” form are continuously updated. The system immediately visualizes the effects of the user changes by resizing the value chains and sorting the related activities. By selecting a value activity in the magnified value chain a graphical display of all the activities that have relationships with the magnified activity are colored. The linked value activities are displayed using two different colors representing the strength of the influence. Clicking on one of the linkage activities brings the “Linkage” form to the front. The linkage form includes information about the magnified activity and a description about the linkage. In Fig. 5, the user selected the value activity “Stock at Depot” in the “Value System” form. This invoked a graphical display using different colors showing all the value activities in the forestry value system that are linked to the selected activity. This function is activated in linkage mode by clicking on a box in the magnified value chain representing a value activity. The selected activity is represented by a blue color.
Activities with strong relationships among performances and costs are displayed using a red color and activities where the influences are not as crucial are displayed using a yellow color. By clicking on one of the linked activities (colored boxes) a pop-up form appear containing a information about the selected linkage. The form, “Linkages” (see Fig. 7), contains a description of the magnified value activity and a description of the selected linked value activity’s cost behavior that is affected by a linkage and interrelationship with the selected magnified activity. The linkages cardinality is also shown. The other activities, that are linked to the magnified value activity are presented in a list. Selecting an activity in this list brings forward information in a description box about the relationship between this activity and the magnified activity in the value system form. The user has the option to create new or delete existing linkages. Selecting the “Create Linkage” option brings a new list on the form showing all the activities that are currently not linked to the magnified activity. The user can select from the list and describe the relationship in the empty “Linkage Description” field. By creating these link changes the graphical representation of the linkages in the “Value System” form is immediately modified. VI. CONCLUSION The value system module is a graphical support for the manager to unfold the major value activities and their linkages in the forestry value system. The prototype DSS provides an
HAAVENGEN et al.: THE VALUE CHAIN COMPONENT IN A DECISION SUPPORT SYSTEM
Fig. 7.
427
Linkages form
intuitive, easy to use interface allowing for direct manipulation of all aspects of such judgements. In turn, this direct manipulation gives the user a better understanding of the consequences of any information about supply and newsprint production that he provides to the module. The value system module was designed to stimulate and assist the manager to make an extended, upstream as well as downstream cost analysis. In such an analysis he reveals the linkages between the suppliers’ value chains that create and deliver inputs through the purchasing division’s activities and influence the company’s value chain. This serves two important purposes; to improve global rationality; and to coordinate and exchange information with factories, suppliers, and transporters. By using the module the manager is also stimulated to search information from research (trade) institutes to enhance the description of the value activities and their linkages. It is not just Porter’s value system that addresses decision maker’s such as the one we have described in this paper Landry [14] noted that while a clear definition of organizational DSS’s is somewhat elusive the concept emphasizes the need for flexible tool kits to support organization-wide decisions which become increasingly problematic as the interactions become more complex. In a working paper Stokke et al. [ 151 noted: . . . current systems for developing competitive strategies share a common weakness. Although they set forth ample theoretical information about the type of analysis to perform none provides practical guidelines for processing large amounts of domain-specific information required for developing such strategies in real life business environments.
We assert that our system addresses and aids this process. The visualization of the value activities’ linkages and their strength, is a tool for the manager to appraise how the effects of the linkages influence the result of his goals-both short term and long term. He appraises in what way his decisions and enterprises exploit the positive effects of the linkages among the value activities and diminish the influence of the negative effects. This process stimulates him to explore future possible development between his division and the other entities in the value system. The value system module gives the manager an aid in constructing mental models of long term goals and perceiving decision making in terms of these goals. Working partnerships in which better coordination of tasks and activities are essential, require increasingly more coordination and mutual adjustments [ 161. As goods and services move along the value-added chain, increasingly a major component of exchange is the exchange of information. The emergence of new ways of packaging or organizing information suggests the importance of information itself as a key variable for analysis. Given our prototype DSS and its accompanying value system module we view our system to be both a mechanism and an enabler that can give rise to a higher quality of information to be used by this manager in his exchange processes. To reveal a linkage that reflects trade-offs among activities and achieve the same overall results, the manager creates the opportunity to lower the total cost of linked activities. We unfolded coordination of activities across organization lines as a means of increasing the manager’s awareness of the dynamics of the value system flow. Frequently the manager needs to expedite a delivery to reduce unanticipated capital demands and offset the quality decay of lumber. Through coordination between the company and the suppliers’ value chains it is possible to benefit both the company and the
IEEE TRANSACTIONS ON ENGINERING MANAGEMENT, VOL. 43, NO. 4, NOVEMBER 1996
428
suppliers. In the coordination activity the manager needs to be more oriented toward the total effects of decision making on all entities, not just his division. We should note here that we did not actually perform a Porter value system analysis per se. Instead, we gave the manager a system wherein he is stimulated to identify certain of his decision environment aspects in a value system form. We are giving him a tool with which we want to stimulate him to identify and describe cost drivers and cost elements and their linkages. Our system at this point is simplistic. However, the manager does not have to be knowledgeable in Porter value system terminology. Following the formulation and initial prototype implementations we revisited the purchasing manager. On these occasions we presented and demonstrated the software. His response was quite enthusiastic to the point that he has requested to employ one of the developers to complete the full DSS. He is at this time actively using the budget and contract prototype and the value system module. A major benefit for the organization using this DSS system together with its instilled value system module is the acquisition and storage of knowledge in the system. This capturing of the manager’s knowledge in the system empowers the organization and gives them an edge against the potential loss of a valued decision maker.
[ 141 J. R. Landry, “Efficient boundaries of organizational decision support systems,’’ in Proc. 27th Annn. Hawaii Int. Con5 on Inform. Syst. Sei.,
1994, pp. 775-784.
[I51 P. Stokke, T. Reve, and T. A. Boyce, “Knowledge engineering and information processing for competitive strategy analysis,” Working Paper no. 10211993, Found. for Res. in Economics and Bus. Admin., Norwegian Sch. Economics and Bus. Admin., Oslo, Nov. 1993. [ 161 E. Christiaanse, “Information as a strategic asset in interfirm relationships,” in Proc. 27th Annu. Hawaii Int. Con5 on Informat. Syst. Sci., 1994, pp. 610-619.
Bj6rn Haavengen received the M S degree in civil engineering and the Ph D degree from the Norwe gian Inatitute of Technology, Trondheim, Norway, in 1989 and 1994, respectively He I5 currently a tesearcher and software engineer for Konigaburg Offshore A S , Konigsburg, Norway
REFERENCES New York: Harper and Row. 1973. R. M. Cyert and J. G. Mar, A Behavioral Theory of the Firm, 2nd ed. Englewod Cliffs, NJ: Prentice-Hall, 1992. J. D. Thompson, “Decision-making, the firm and the market,” in New’ Perspectives in Organization Research, W. W. Cooper, H. J. Leavilt, and M. W. Shelly 11, Eds. New York: Wiley, 1964. G. L. Frazier and R. C. Rody, “The use and influence strategies in interfirm relationships in industrial product channels,” J. Market., vol. 5 5 , pp. 52-69, Jan. 1991. R. A. Friedman and J. Podolny, “Differentiation of boundary spanning roles; labor negotiations and implications for role conflict,” Harvard Bus. School Working Paper, July 1991. M. Zeleny, Multiple Criteria Decision Making. New York: McGrawHill, 1982. J. F. Rockart, “Chief executives define their own data needs,” Havvnrd Bus. Rev., pp. 81-93, Mar.-Apr. 1979. C. V. Bullen and J. F. Rockart, “A primer on critical success factors,” Center for Information Syst. Res. Working Paper no. 69, MIT Sloan Sch. Manage., Cambridge, MA, June 1981. M. J. Shapiro and G. M. Bonham, “Cognitive process and foreign policy decision-making,” Int. Studies Quart, vol. 17, no. 2, pp. 147-174, 1973. A. M. Fuglseth, Beslutningsst@tte:Metode .for diagnose av ledernes informasjons ug siluasjonsopp~atninger,Avhandling for graden dr.oecon., Norges HandelshGgskolc, 1989. A. M. Fuglseth and C. B. Stabell, “Capture, representation, and diagnosis
[ 11 H. Mintzberg. The Nature ofManagerial Work.
121 [31
161 171
of user information perseption,” Working Paper, The Norwegian Sch
Manage., 1983. K. D. Wall, “A model of decision making under bounded rationality,” J. Economic Behav. and Organiz., vol. 21, pp. 331-352, 1993. M. E. Porter, Competitive Advantage. New York: Free Press, 1985.
Dag H. Olsen received the M.S. degree in physics from the University of Oslo, Norway, the M.B.A. degree from Texas A&M University, College Station, and the Ph.D. degree from the Norwegian Institute of Technology, Trondheim, in 1986, 1989, and 1995, respectively. He is currently an Associate Professor at Agder College, Kristiansand, Norway. His current research interests are in the use of information technology in management and organizations, decision support systems, and decision research.
James A. Sena received the Ph.D. degree in management information systems from the University of Kentucky, Lexington. He is currently a Professor of Business Computing at California Polytechnic State University, San Luis Obispo. Prior to that, he was a professor at the University of Louisville, the University of Houston, and Texas A&M University. His current research interests focus on decision support systems, groupware and team performance, and business process improvement, He has published papers in the European Journal of Purchasing and Supply, the Journal of Information Technology, the Journal of Applied Behavioral Science, and the Oil and Gas Tax Quart1?rZy.
