Willem J. H. van Groenendaal

On the assessment of economic risk : Factorial design versus Monte Carlo methods

Abstract The feasibility of large investment projects (such as gas transmission and power system projects) has many aspects. Usually, this problem cannot be modeled as a single optimization problem; instead, the multiple aspects (demand, supply, prices, investment costs) are modeled separately. Each aspect may require a large, nonlinear submodel. The results of such a submodel can often be summarized by one or a few variables, which combine all the submodels information; for example, total demand is the sum of the demand per customer type, each type being modeled separately. Traditionally, the feasibility of the investment project is then judged by combining the results of the various submodels for the ‘base case’ values of all model inputs. This base case information, however, is not sufficient for the decision makers; they also like to know the economic risk they are taking. To assess this risk on the project level (Hertz, D. B., Risk analysis in capital investment. Harvard Business Review, 1964, 95–106) developed a method known as risk analysis. This method is based on the estimated probability distribution of a projects net present value (NPV). This distribution is obtained by introducing distributions for the model inputs. The projects economic risk is then. expressed as the probability of a negative NPV exceeding a critical value (say) α. Nowadays this approach is becoming popular, because many software packages (such as @RISK and Crystal Ball) facilitate such a risk analysis. Although Hertzs risk analysis is appealing, it has a number of theoretical and practical flaws, which may lead to wrong conclusions. These flaws are discussed in this paper. From a modelling point of view, Hertzs risk analysis is similar to analysing the technological or operational risk of an investment. However, economic risk and technological risk are different concepts that require different analyses. In this paper these differences are discussed and it is shown that Hertzs risk analysis does not measure what is normally meant by a projects economic risk. Furthermore, the information requirements for the application of risk analysis to large investment projects are formidable; this makes the results of Hertzs investment analysis unreliable. Less information is required by sensitivity analysis based on the statistical design of experiments (such as 2k−P designs); this analysis is more robust, and leads to results that better satisfy the information needs of decision makers.

Measuring the quality of publications: new methodology and case study

In practice, it is important to evaluate the quality of research, in order to make decisions on tenure, funding, and so on. This article develops a methodology using citations to measure the quality of journals, proceedings, and book publishers. (Citations are also used by the Science and Social Science Citation Indexes, published by the Institute for Scientific Information (ISI), but these Indexes do not cover proceedings, books, and certain journals.) The novel methodology uses statistical sampling, bootstrapping, and classification. This methodology is applied to the field of Information Systems. In this case-study, class-1 turns out to consist of three journals - MIS Quarterly, Management Science, and Communications of the ACM - and two proceedings - VLDB and SIGMOD. The class-1 publishers are Springer, Wiley, and Addison-Wesley. Moreover, hundreds of other journals etc. are classified into a small number of classes.

Deterministic versus stochastic sensitivity analysis in investment problems: An environmental case study

Sensitivity analysis in investment problems is an important tool to determine which factors can jeopardize the future of the investment.Information on the probability distribution of those factors that affect the investment is mostly lacking.In those situations the analysts have two options: (i) apply a method that does not require knowledge of that distribution, or (ii) make assumptions about the distribution.In both approaches sensitivity analysis should result in practical information about the actual importance of potential factors.For approach (i) we apply statistical design of experiments (DOE) in combination with regression analysis or meta-modeling.For approach (ii) we investigate five types of relationships between the model output and each individual factor; Pearsons p, Spearmans rank correlation, and location, dispersion, and statistical dependence.We introduce two distribution types popular with practitioners: uniform and triangular.In an environmental case study both approaches identify the same factors as important.

Two-stage versus sequential sample-size determination in regression analysis of simulation experiments

SYNOPTIC ABSTRACTMany simulation models have responses with variances that vary with the inputs. Then consider the number of observations (simulation runs, replications) per combination of inputs. These numbers can be selected such that the variances of the average responses become approximately equal: system variants with high variability are simulated more often. These average responses can be analyzed through a regression (meta)model. The regression parameters can be estimated through Weighted Least Squares (WLS). WLS becomes identical to Ordinary Least Squares (OLS) applied to the average responses. Because the variances are unknown, they are estimated by repeating the runs with different random numbers. The estimated variances yield the number of runs required to obtain approximately equal variances per average response. Two rules for selecting the required number of runs are presented, namely a two-stage and a sequential rule. These stopping rules are first formalized and analyzed; next they are fur...

Solving misfits in ERP implementations by SMEs

The gap between the organizational needs and the extent to which an ‘off‐the‐shelf’ enterprise resource planning (ERP) system can meet these is called a misfit. A framework is developed to distinguish actual from perceived misfits. This is used to analyse the ERP implementation at four small‐sized and medium‐sized enterprises. The results show that they prefer to adjust the ERP system to their business processes when needed but often unnecessarily change the system to solve perceived misfits. The framework is a first step to prevent this unnecessary work in the future.

Regression metamodels and design of experiments

Simulation models are often used to support decision making for problems with uncertain inputs and parameters. Three types of models are used: deterministic, risk, and uncertainty models. Risk models are popular with researchers, but can be used only when the joint probability distribution of the inputs and parameters is known. In many real-life situations, however, this is not the case. Uncertainty models are too restrictive for real-life situations. Therefore deterministic models are then used. The sensitivity of the results is often analyzed by changing one factor at a time or by simulating a few scenarios. This paper, however, shows that in case of uncertainty it might be better to apply design of experiments (DOE) in combination with regression metamodels.

Assessing demand when introducing a new fuel: Natural gas on Java

The Indonesian government is investing in a gas transmission system on Java. For the evaluation of this investment a forecast of the demand for natural gas by the manufacturing sector is needed. To obtain this forecast the manufacturing sector is divided into subsectors according to energy use in production processes. On the level of production processes the opportunities for natural gas are based on net present value evaluations of its future benefits in production. This results in the desired fuel mix for manufacturing subsectors, from which the gas intensity ratios per subsector for existing production and new investments are calculated. Gas demand can then be forecast by combining the gas intensity ratios with subsectoral (growth in) gross value-added. This approach leads to a flexible forecasting tool that can readily account for changes in economic structure and energy prices, as encountered by rapidly developing economies.

A user's view on the electronic library

The electronic library offers researchers the opportunity to consult more information sources than ever before. The times of long visits to the library to browse through books and journals belong in the past for many of us. In addition, the personal relationships with experienced library staff who knew your interests has disappeared. This has been replaced by search profiles that supply you with regular updates on new publications in your field of research. The electronic library is, however, more than a digitalized version of the old library. Searching has become easy and quick, and the number of databases that can be accessed has increased significantly. For most users the library is, however, no longer the only, or even the main, road along which information becomes available. It is used in combination with the many sources that have emerged along the electronic highway. The main problem for the user is how to manage all these different resources. Although the library information system is relatively easy to use and shows a large number of user-friendly extras, it is not what he/she really needs. What the user really needs is intelligent assistance in using this combination of local and remote information sources. However, nothing is for free. The question is how are we going to pay for all these new opportunities in a time of diminishing budgets. The answer may be to make full use of the electronic library.

Supporting Strategic Decision Processes

Abstract Decision support systems often focus on those aspect of problem solving that require choice. Decision making as a process is often neglected. Strategic decisions often require many a long period of time, and supporting the decision process is more important than supporting the search for an optimal solution of the problem, especially when optimal solutions only exist for some subproblems. A problem with these features is the decision to invest in gas transmission on the island of Java. The opportunities to support this process are analyzed, and a specialized DSS is designed that allows long term support. Along with the design of the DSS, its restrictions for application are determined.

Control, coordination and conflict on international commodity markets

Abstract The effects of price stabilization policies have been investigated for both theoretical and empirical models. The institutional construct for international commodity markets is that a buffer stock manager employs a band width rule or a price adjustment rule to stabilize the world market price. In those investigations it is assumed that the other market participants do not react to the stabilization activities of the buffer stock manager. This paper describes an international commodity market as a difference game between buffer stock manager, producing countries and consuming countries and uses an empirical model for the world cocoa market to analyse the effects. The feedback Nash behavioural equilibrium for this game is compared with the optimal control outcome which ignores the strategic behaviour in producing countries and consuming countries. It is found that producers and consumers engage in storage activities which have a negative effect on the stabilization efforts of the buffer stock manager but which decrease the operating costs of the buffer stock, decrease the revenues of the producers and decrease the costs of consumers.