A KNOWLEDGE-BASED DSS FOR MANAGERIAL PROBLEM DIAGNOSIS

A knowledge-based system supporting managerial problem diagnosis is described. The system provides the capability to monitor values of selected variables for problem situations. When problems are located, a list of problem symptoms is delivered to a problem processor for structuring and diagnosis. Problem structuring is based on a combination of concepts from expert systems and structural modeling. User assertions about cause-effect relationships between pairs of variables are maintained in a semantic network. Problem diagnosis uses the relationships in the semantic network to construct causation trees, the branches of which represent potential explanations of the problem symptoms. Mathematical models are constructed based on causation-tree branches, and values from the data base are used to test whether the model confirms the diagnosis. If so, the source of the problem has been located and it is then up to the user to resolve the problem. If the model fails to explain the problem, the model apparently is deficient and the user may perform “what if…” type scenarios in attempts to improve the model and search for problem causes. Realistic applications in the accounting and health care areas are discussed.     

Applications and Implementation DECISION SUPPORT SYSTEMS: AN EMPIRICAL ASSESSMENT FOR DECISION MAKING

A laboratory experiment examined the effects of applying decision support system (DSS) technology to decision making in ill-structured problem environments under varying information conditions. Marketing executives participated in the experiment which investigated the effects of DSS availability, DSS training, and data availability on dependent variables that included: (1) the number of alternatives considered by a subject during decision making, (2) the period of time spent by a subject to complete the decision-making process, (3) the subject's perceived confidence in the decisions he or she had made, (4) the amount of data considered by a subject's during decision making, (5) the individual subject's decision processing, and (6) the subject's performance overall. Our results indicate that all three factors significantly affect the number of alternatives considered by subjects during the decision-making process. We therefore suggest that DSS training be coordinated with decision training in order to realize the potential of DSSs as described in the DSS literature.     

A NOTE ON HOLDING COSTS AND LOT-SIZE ERRORS

This note examines the sensitivity of the basic economic-order-quantity inventory model to lot-size errors when holding costs are assumed to be a strictly increasing (though not necessarily linear) function of average inventory. In particular, we show that the penalty associated with ordering either too much or too little is a function not only of the size of the error but of the shape of the holding-cost curve as well. We demonstrate that, under certain conditions, even relatively small lot-size errors can be extremely costly.     

EMPIRICAL AND JUDGMENT-BASED SALES-FORCE DECISION MODELS: A COMPARATIVE ANALYSIS*

Multiple-predictor empirical and single-predictor judgment-based models represent the two basic types of sales-force decision models (SDMs). The important similarities and differences between the two modeling approaches are identified and discussed. The SDM literature is reviewed to determine how each model type has been applied and to evaluate previous attempts at model validation. A research study designed to compare the normative sales-call allocations generated from each modeling approach is presented. Study results indicate that empirical and judgment-based SDMs would produce similar normative guidelines. The implications of this finding are discussed.     

CATEGORY WIDTH COGNITIVE STYLE AS A FACTOR IN ACCOUNTANTS' PERCEPTIONS OF ACCOUNTING INFORMATION

This paper reports the findings of a laboratory experiment designed to investigate the relationship of category width (CW) cognitive style with accountants' perceptions of accounting information. Subjects drawn from large accounting firms in Sydney, Brisbane, and Melbourne, Australia, were classified into broad, medium, and narrow categories following the test devised by Pettigrew [19]. Subjects were requested to state their level of confidence in decisions they had made after receiving (1) conventional accounting information or (2) conventional accounting information and human resources accounting (HRA) information using a one-group pretest-posttest design. The results indicated a significant relationship between CW cognitive style and the accountants' confidence in their decisions. Furthermore, CW cognitive style moderated the accounting-information/decision-making relationship.     

THE IMPACT OF EXTREME OBSERVATIONS ON SIMPLE FORECASTING METHODS*

In general linear modeling, an alternative to the method of least squares (LS) is the least absolute deviations (LAD) procedure. Although LS is more widely used, the LAD approach yields better estimates in the presence of outliers. In this paper, we examine the performance of LAD estimators for the parameters of the first-order autoregressive model in the presence of outliers. A simulation study compared these estimates with those given by LS. The general conclusion is that LAD does not deal successfully with additive outliers. A simple procedure is proposed which allows exception reporting when outliers occur.     

A LOWER BOUND FOR THE SINGLE-LEVEL DYNAMIC HORIZON LOT-SIZING PROBLEM

Conventional production planning methods assume the existence of a medium- or longrange demand horizon. However, demand usually is known over a much shorter range; scheduling decisions must be made within this “decision window,” which rolls forward in time. This paper presents a new lower bound for lot-sizing heuristics in a rolling-horizon framework and compares it to the well-known Wagner-Whitin bound. The new bound indicates heuristic schedules that have costs close to the optimum. Rolling-horizon schedule costs are compared to corresponding static-horizon schedule costs (assuming the whole horizon is known in advance), using the ratio of decision-window size to the natural order cycle as a parameter. For values below unity, the rolling-horizon policy is significantly more costly. For values above one, the two policies have similar costs and actually converge as the parameter value increases.     

A DISCRETE DYNAMIC OPTIMIZATION MODEL FOR MADE-TO-ORDER COST ANALYSIS

A model is presented that yields optimal production rates for a firm producing a contracted order. The model is unique in that it considers the influence of production rate and learning on total program cost. An application to the specific characteristics of two military production programs is presented. As demonstrated by the application, models of this type may be used as decision-making tools when negotiating the cost impact of contract modifications.     

COST INCREASES DUE TO DEMAND UNCERTAINTY IN MRP LOT SIZING

This paper analyzes the cost increases due to demand uncertainty in single-level MRP lot sizing on a rolling horizon. It is shown that forecast errors have a tremendous effect on the cost effectiveness of lot-sizing techniques even when these forecast errors are small. Moreover, the cost differences between different techniques become rather insignificant in the presence of forecast errors. Since most industrial firms face demand uncertainty to some extent, our findings may have important managerial implications. Various simulation experiments give insight into both the nature and the magnitude of the cost increases for different heuristics. Analytical results are developed for the constant-demand case with random noise and forecasting by exponential smoothing. It is also shown how optimal buffers can be obtained by use of a simple model. Although the analysis in this paper is restricted to simplified cases, the results merit further consideration and study. This paper is one of the first to inject forecast errors into MRP lot-sizing research. As such it attempts to deal with one of the major objections against the practical relevance of previous research in this area.     

A SIMULATION ANALYSIS OF THE JAPANESE JUST-IN-TIME TECHNIQUE (WITH KANBANS) FOR A MULTILINE,MULTISTAGE PRODUCTION SYSTEM

The Japanese “just-in-time with kanban” technique reduces in-process inventory to absolute minimal levels, in concert with the Japanese belief that inventory is an unnecessary evil. Due to the success of Japanese firms that employ this type of system, American firms would like to import this technique and emulate Japanese successes. But this Japanese success may be attributable not only to the just-in-time with kanban technique but also to the production environment in which the technique is employed. This paper simulates the just-in-time with kanban technique for a multiline, multistage production system in order to determine its adaptability to an American production environment that might include such characteristics as variable processing times, variable master production scheduling, and imbalances between production stages. The results have practical implications for those firms considering adoption of the Japanese technique.