MVLSC HomeIssue Contents

Preface: Fuzzy Decision Making in Risk Management
Cengiz Kahraman

This special issue covers several representative applications of decision-making in risk management. It contains seven original research and application-oriented papers covering different areas of decision-making under risk. Most of them are the substantially extended versions of the papers selected from 150 contributions presented at the 2nd International Conference on Risk Analysis and Crisis Response (RACR-2009), organized by Risk Analysis Council of China Association for Disaster Prevention (RAC), and held at Peking University, China, in October 19–21, 2009.

Decision-making is the cognitive process of selecting a course of action from multiple alternatives. Fuzzy set approaches to decision-making are usually most appropriate when human evaluations and the modeling of human knowledge are needed. Risk is the chance that an undesirable event will occur and the consequences of all its possible outcomes. Another definition for risk is that it is a measure of the probability and consequence of not achieving a defined goal. Most people agree that risk involves the notion of uncertainty. However, when risk is considered, the consequences or damage associated with the event occurring must also be considered. Risk is not always easy to evaluate, since the probability of occurrence and the consequence of occurrence are usually not directly measurable parameters and must be estimated by statistical or other procedures.

Risk management is the identification, assessment, and prioritization of risks followed by coordinated and economical application of resources to minimize, monitor, and control the probability and/or impact of unfortunate events or to maximize the realization of opportunities. The components of Risk Management are planning, assessment, handling, and monitoring. Risk planning is the process of developing and documenting an organized, comprehensive, and interactive strategy and methods for identifying and analyzing risk issues, developing risk handling plans, and monitoring how risks have changed. Risk assessment is the process of identifying and analyzing program areas and critical technical process risks to increase the likelihood of meeting cost, performance, and schedule objectives. Risk identification is the process of examining the program areas and each critical technical process to identify and document the associated risk. Risk analysis is the process of examining each identified risk issue to estimate the likelihood of a risk and predict its impacts. Risk handling is the process that identifies, evaluates, selects, and implements one or more strategies in order to set risk at acceptable levels given program constraints and objectives. Risk monitoring is the process that systematically tracks and evaluates the performance of risk handling actions against established metrics throughout the acquisition process and provides inputs to updating risk handling strategies, as appropriate.

The failure data available are usually scarce and often accompanied with a high degree of uncertainty. For this reason the use of conventional probabilistic risk assessment methods may not be well suited. Risk management should take into account the vagueness and uncertainty inherent in risk and provide a good assessment based upon experts judgments. Fuzziness is the uncertainty resulting from vagueness. Most natural language descriptors are vague and somewhat uncertain, rather than precise.

In the first paper of this issue, a linguistic-valued lattice implication algebra approach for risk analysis is presented. The approach is based on a 10-element lattice with five linguistic hedges and two antonyms basic words. The proposed method better expresses and handles both comparable and incomparable information in risk analysis domains. The operations and special properties of the 10 evaluating linguistic values are discussed by their lattice implication algebra, which are suitable for expressing semantics of the 10 evaluating linguistic values as well as their logical characters. The reasoning and operation are directly acted by the evaluating linguistic values in the risk analysis process. An illustration example shows the proposed approach seems more effective for risk analysis under a fuzzy environment with both comparable and incomparable linguistic values.

The second paper proposes a fuzzy multi-criteria decision making methodology for the selection of waste-water treatment plant location among alternatives. The methodology is based on the analytic hierarchy process (AHP) under fuzziness. It allows the evaluation scores from experts to be linguistic expressions, crisp, or fuzzy numbers. In the application of the proposed methodology, the best location is determined for Malatya in Turkey.

The next paper considers the accuracy of forecasting models based on statistical (stochastic) methods sometimes called hard computing and soft methodology based on soft or granular computing. A new method for finding the forecasting horizon within which the risk is minimal is also presented. To evaluate the risk, some methods are presented based on the analysis of forecast errors by applying the exponential smoothing concept.

The fourth paper proposes a hybrid risk evaluation model for real estate investments which integrates Fuzzy rule based systems (FBRS) with Fuzzy Analytic Hierarchy Process (FAHP). In the proposed hybrid model, FAHP is used to evaluate the real estate quality analytically with respect to the judgments of experts. Then, the obtained quality score is utilized as one the inputs of FRBS method which uses fuzzy IF-THEN rules defined by the experts. The rule base of the hybrid model combines the factors of age and quality of a real estate with general economical factors of the market. The proposed model is applied to determine the prices of three different real estate in Istanbul based on quality of house, size and risk factor.

The fifth paper outlines the research which develops a knowledge-based method that makes full use of previous knowledge to perform a comprehensive forecast of the risk of avian influenza and generate reliable warning signals for a specific region at a specific time. The method contains a risk estimation model and a knowledge-based prediction method using fuzzy logic and case-based reasoning (CBR) to generate timely early warnings to support decision makers to identify underlying vulnerabilities and implement relevant strategies. An example is presented that illustrates the capabilities and procedures of the proposed method in avian influenza early warning systems.

In the next paper, a risk assessment methodology is developed to calculate the total risk magnitude of a project by taking the risk magnitude of each factor into account. The mentioned parameters are commonly used to calculate risk magnitude in the existing studies in literature. The developed methodology is based on a multi-criteria decision making method and handles risk likelihood and risk severity of each risk factor in construction project that lead to a project failure. The main structure of the proposed methodology consists of analytic hierarchy process (AHP) integrated with a linguistic evaluation procedure. The proposed algorithm is applied to the risk analysis of a housing project to illustrate how a real case implementation of the methodology can be achieved in practice.

The last paper of this issue considers six main risk factors: executive risks, organizational risks, project management risks, technical risks, decision making risks, and functional risks. Having “evaluation of the failure risk of a SAP/ R3 implementation project” as the main objective, possible inter and inner relationships/dependencies between the main factors are considered and an Analytic Network Process approach is proposed to the model. Decision makers usually are unable to be explicit about their preferences due to the fuzzy nature of the comparison process. So, expert judgments are matched with fuzzy numbers in this study.

I hope that this special issue will serve as a useful source of ideas, techniques, and methods for further research in the applications of fuzzy sets to risk measurement and management. I am grateful to the referees whose valuable and highly appreciated works contributed to the selection of the high quality papers published in this special issue. My sincere thanks go to Profs. Ivan Stojmenovic and Dan Simovici, the editors-in-chief, who were highly instrumental in bringing this project to its fruitful completion.

Full Text (IP)