An integrative evaluation framework for intelligent decision support systems

Evaluation of the overall effectiveness of decision support systems (DSS) has been a research topic since the early 1980s. As artificial intelligence methods have been incorporated into systems to create intelligent decision support systems (IDSS), researchers have attempted to quantify the value of the additional capabilities. Despite the useful and relevant insights generated by previous research, existing evaluation methodologies offer only a fragmented and incomplete view of IDSS value and the contribution of its technical infrastructure. This paper proposes an integrative, multiple criteria IDSS evaluation framework through a model that links the decision value of an IDSS to both the outcome from, and process of, decision making and down to specific components of the IDSS. The proposed methodology provides the designer and developer specific guidance on the intelligent tools most useful for a specific user with a particular decision problem. The proposed framework is illustrated by evaluating an actual IDSS that coordinates management of urban infrastructures.     

Reusable components in decision tree induction algorithms

We propose a generic decision tree framework that supports reusable components design. The proposed generic decision tree framework consists of several sub-problems which were recognized by analyzing well-known decision tree induction algorithms, namely ID3, C4.5, CART, CHAID, QUEST, GUIDE, CRUISE, and CTREE. We identified reusable components in these algorithms as well as in several of their partial improvements that can be used as solutions for sub-problems in the generic decision tree framework. The identified components can now be used outside the algorithm they originate from. Combining reusable components allows the replication of original algorithms, their modification but also the creation of new decision tree induction algorithms. Every original algorithm can outperform other algorithms under specific conditions but can also perform poorly when these conditions change. Reusable components allow exchanging of solutions from various algorithms and fast design of new algorithms. We offer a generic framework for component-based algorithms design that enhances understanding, testing and usability of decision tree algorithm parts.     

Concurrent optimization in designing for logistics support

Concurrent Engineering has always been an important field within military industry and is gaining recognition within commercial industry as well. Studies have suggested that long-range research efforts need to be directed to the area of integrated logistics. In order to satisfy the overall logistics objectives, product designs must incorporate consideration of the impact that design decisions have on operations, maintenance, transportation and supply. This paper presents a model that carries out a concurrent optimization of a product design and its associated manufacturing and logistics support systems. We offer this model as a framework for decision support of concurrent engineering activities.     

Control of arrivals and departures in a state-dependent input-output system

At each decision epoch, an offer for a unit to either enter or leave a system is received. These offers arrive according to a Poisson process. With each offer is associated a value revealed upon the arrival of the offer. The distribution of the value of the offer is given and is a function of what kind of offer is received (enter or leave). The decision to accept or reject and offer is allowed to depend on the current state and the current value received. The objective is to maximize the expected discounted difference between the sum of the accepted output offers and the sum of the accepted input offers. The key result of the paper that under various conditions, the decision to accept or reject an offer depends on whether or not its value is above or below, respectively, a critical value that depends on the state of the system.     

Investment decisions in the rent-to-own industry in the absence of inventory

This study addresses the product investment decision faced by firms in the rent-to-own industry. In this setting, a customer arrives according to a random process and requests one unit of a product to rent (and eventually own should he/she choose to make all the required payments). At the time of request, if the product is available in inventory, the firm enters into a contractual agreement (by accepting the customer's offer) and rents the merchandise. More interesting and the case considered here, if the requested item is not in inventory, the firm must decide whether to purchase the item in order to rent it out or to simply reject the request. The customer's offer specifies the desired maximum contract length and the payment frequency—from which the firm determines the fixed periodic payment charged. The firm makes its investment decision based on the characteristics of the offer as well as those of the product (eg, initial and resale values, useful life and carrying costs) in essence performing a complicated cost benefit analysis. An extension is also considered whereby instead of simply rejecting the request the firm can adjust the required payment amount. Dynamic programming techniques are used to address the problem and to solve for the firm's optimal decision.     

Trade deals and/or on-package coupons

The selection of either a pull or a push price promotion has mainly been investigated in contexts where manufacturers offer deals to consumers at the time of purchase or offer trade deals to retailers. This paper extends this framework to where manufacturers can offer either trade deals or rebate-like promotions to consumers such as on-pack coupons that stimulate the first and second purchases or a combination of the two promotion vehicles. It is demonstrated that the decision to implement either of the three promotion options critically depends, among other factors, on the percentage of first-time buyers who redeem their coupons at the second purchase. Particularly, a necessary condition to simultaneously offer both a trade deal and coupons is to have a positive coupon redemption rate. When possible, manufacturers prefer on-pack coupons over trade deals to take advantage of slippage and to further increase the overall demand via coupon-induced repeat purchase. Manufacturers are more likely to take the lion’s share of channel profits.     

Hit or miss: A decision support system framework for signing new musical talent

In the music industry, the process of signing new musical talent is one of the most complex decision-making problems. The decision, which is generally made by an artist and repertoire (A&R) team, involves consideration of various quantitative and qualitative criteria, and usually results in a low success rate. We conducted a series of mental model interviews with the aim of developing a decision support framework for A&R teams. This framework was validated by creating a decision support system that utilises multi-criteria decision analysis to support decision-making. Our framework and subsequent implementation of the decision support system involving decision rule and weighted sum methods show an improvement in the ability to analyse and decide on greater amounts of talent. This paper serves as a building block for developing systems to aid in this complex decision-making problem.     

Optimal stopping problem with double reservation value property

This paper considers an optimal stopping problem with uncertain recall probability where some recall cost must be paid to accept a past offer. Recall cost is an important factor leading to value deterioration of an offer. However, it is assumed in all the conventional models so far presented that the value of a past offer does not change with time, and its original value is guaranteed even if the past offer is recalled at any time in the future. In the model of this paper, the value of a past offer decreases due to the recall cost. The purpose of this paper is to propose a model with such a recall cost factor and uncertain recall probability, and to examine the properties of its optimal decision rule. One of the most distinctive results in this study is that the optimal decision rule may have a seemingly counter-intuitive property, called the DRV (Double Reservation Value) property: the critical point of whether or not to stop the search by accepting an offer is not a single entity. An interpretation of why the property may appear in this model will be given. Furthermore, this paper examines the necessary and sufficient conditions required for the property to appear, and demonstrates that the recall cost is an essential factor to the appearance of the property. To illustrate the importance of considering the recall cost factor and uncertain recall probability, this paper also provides some practical implications to a decision maker confronted with a problem of M&A (Merger & Acquisition), using an actual case of the Ford–Daewoo deal.     

Mathematical Programming Approaches to Sensitivity Calculations in Decision Analysis

A general framework for sensitivity analysis of discrete multi-criteria decision problems with or without uncertainty has been developed. The framework results in the need to solve very many mathematical programmes and hence is being implemented using parallel programming techniques. The method is illustrated with reference to a simple decision tree problem.     

Decision making using multiple models

Many real world business situations require classification decisions that must often be made on the basis of judgment and past performance. In this paper, we propose a decision framework that combines multiple models or techniques in a complementary fashion to provide input to managers who make such decisions on a routine basis. We illustrate the framework by specifically using five different classification techniques – neural networks, discriminant analysis, quadratic discriminant analysis (QDA), k-nearest neighbor (KNN), and multinomial logistic regression analysis (MNL). Application of the decision framework to an actual retail department store data shows that it is most useful in those cases where uncertainty is high and a priori classification cannot be made with a high degree of reliability. The proposed framework thus enhances the value of exception reporting, and provides managers additional insights into the phenomenon being studied.     

Pairwise-comparison methods in multiple objective programming,with applications in a long-term energy-planning model

We propose to use pairwise comparisons within the framework of ideal-point or reference-point methods for multi-objective programming. The decision makers are requested to estimate the ratios which are acceptable for deviations from the ideal vector. Thereafter, we seek the nearest feasible solution using the Tchebycheff norm. In this paper we sketch the pairwise-comparison methods, some aspects of magnitude scaling, and the ideal-point methods. We show the results of our numerical experiments in long-term energy planning with nine objective functions. Finally, we present a preliminary evaluation of the combined method on the basis of its possible contribution to interactive decision analysis.     

Some issues in mathematical modelling of multiple criteria decision-making problems

Models for analyzing and solving multiple criteria decision-making (MCDM) problems are difficult to evaluate and compare, because they are intended for diverse orderings of a set of feasible alternatives. These models are based on a variety of assumptions about the decision maker's preferences and use different types of preference information. In this paper, a conceptual framework is developed for evaluating and comparing discrete alternative MCDM models available for a given decision situation. The procedure employed in the framework guides the user through an analysis of the decision situation making it possible for a decision maker or analyst to select the most appropriate MCDM model from among several alternative feasible models.     

An exploratory framework for handling the complexity of mathematical problem posing in small groups

The paper introduces an exploratory framework for handling the complexity of students’ mathematical problem posing in small groups. The framework integrates four facets known from past research: task organization, students’ knowledge base, problem-posing heuristics and schemes, and group dynamics and interactions. In addition, it contains a new facet, individual considerations of aptness, which accounts for the posers’ comprehensions of implicit requirements of a problem-posing task and reflects their assumptions about the relative importance of these requirements. The framework is first argued theoretically. The framework at work is illustrated by its application to a situation, in which two groups of high-school students with similar background were given the same problem-posing task, but acted very differently. The novelty and usefulness of the framework is attributed to its three main features: it supports fine-grained analysis of directly observed problem-posing processes, it has a confluence nature, it attempts to account for hidden mechanisms involved in students’ decision making while posing problems.     

A framework for sensitivity analysis of decision trees

In the paper, we consider sequential decision problems with uncertainty, represented as decision trees. Sensitivity analysis is always a crucial element of decision making and in decision trees it often focuses on probabilities. In the stochastic model considered, the user often has only limited information about the true values of probabilities. We develop a framework for performing sensitivity analysis of optimal strategies accounting for this distributional uncertainty. We design this robust optimization approach in an intuitive and not overly technical way, to make it simple to apply in daily managerial practice. The proposed framework allows for (1) analysis of the stability of the expected-value-maximizing strategy and (2) identification of strategies which are robust with respect to pessimistic/optimistic/mode-favoring perturbations of probabilities. We verify the properties of our approach in two cases: (a) probabilities in a tree are the primitives of the model and can be modified independently; (b) probabilities in a tree reflect some underlying, structural probabilities, and are interrelated. We provide a free software tool implementing the methods described.

     

Risk analysis and decision theory: A bridge

The risk-triplet approach pioneered by Kaplan and Garrick is the keystone of operational risk analysis. We perform a sharp embedding of the elements of this framework into the one of formal decision theory, which is mainly concerned with the methodological and modeling issues of decision making. The aim of this exercise is twofold: on the one hand, it gives operational risk analysis a direct access to the rich toolbox that decision theory has developed, in the last decades, in order to deal with complex layers of uncertainty; on the other, it exposes decision theory to the challenges of operational risk analysis, thus providing it with broader scope and new stimuli.     

Mathematical programming and the sensitivity of multi-criteria decisions

We report on the current state of a project whose aim is to implement a framework for sensitivity analysis in Multi-Criteria Decision Making (MCDM). The framework is largely based on mathematical programming. Due to the potentially large number and nature of the mathematical programmes, it is far from trivial to provide solutions to all of them in acceptable computing times. The challenge is even greater when we recognize that much decision analysis is performed in the context of decision conferences where any sensitivity analysis needs to be conducted in near real time (preferably) on a PC. We present a parallel processing approach to this challenge and point to some of the difficulties still to be resolved. Preliminary results obtained on a network of transputers are discussed.     

Decentralized operation strategies for an integrated building energy system using a memetic algorithm

The emerging technology in net-zero building and smart grids drives research moving from centralized operation decisions on a single building to decentralized decisions on a group of buildings, termed a building cluster which shares energy resources locally and globally. However, current research has focused on developing an accurate simulation of single building energy usage which limits its application to building clusters as scenarios such as energy sharing and competition cannot be modeled and studied. We hypothesize that the study of energy usage for a group of buildings instead of one single building will result in a cost effective building system which in turn will be resilient to power disruption. To this end, this paper develops a decision model based on a building cluster simulator with each building modeled by energy consumption, storage and generation sub modules. Assuming each building is interested in minimizing its energy cost, a bi-level operation decision framework based on a memetic algorithm is proposed to study the tradeoff in energy usage among the group of buildings. Two additional metrics, measuring the comfort level and the degree of dependencies on the power grid are introduced for the analysis. The experimental result demonstrates that the proposed framework is capable of deriving the Pareto solutions for the building cluster in a decentralized manner. The Pareto solutions not only enable multiple dimensional tradeoff analysis, but also provide valuable insight for determining pricing mechanisms and power grid capacity.     

Expanding competence sets for the consumer decision problem

The framework of competence set analysis provides a new approach to complement the existing models for the consumer decision problem. Since it is essential for the firm to improve the competence set of its product or service to fully address the consumer's truly needed benefits, the effective expansion of competence sets plays an important role in marketing reality. The previous studies regarding competence set expansion have thrown light on the tree expansion processes, but forest learning is more suitable for the acquisition of product benefits than tree learning. Thus, this study looses the assumption of tree learning and conducts forest learning to design an effective expansion program. In addition, this study explores a more general problem involving intermediate attributes, compound benefits, and experiential effects. An algorithm is also provided for effective expansion of competence sets in consumer decision analysis.     

On the nearest parametric approximation of a fuzzy number

Many nearest parametric approximation methods of fuzzy sets are proposed in the literature. It is clear that the specific approximations may lead to the loss of information about fuzziness. To overcome this problem, most of these methods rely on the minimization of the distance between the original fuzzy set and its approximation. But these approximations mostly are not flexible to the decision maker's choice. Hence, in this paper, we offer a parametric fuzzy approximation method based on the decision maker's strategy as an extension of trapezoidal approximation of a fuzzy number. This method comprises the selection of the form of the parametric membership function and its evaluation.     

Using competence sets to analyze the consumer decision problem

This study is intended to provide a different approach to complement the existing consumer decision models (CDMs). Based on the concept of habitual domains and competence sets, we supply a framework for helping a firm in expanding the benefits of its products to fully address the consumer’s needs. According to the features of consumers’ decision making, we use challenging problem types to explore extensive problem solving, fuzzy problem types for limited problem solving, and routine and mixed routine problem types for routine problem solving. In addition, several useful indexes are established using fuzzy measures in this study, including the possibility of successfully appealing to consumers, the degree of consumer satisfaction, the degree of compatibility, and the degree of uniqueness. These indexes can be a decision support for implementing competence set analysis in practical applications. Finally, an empirical study on children’s apparel was conducted to show the applicability and feasibility of our proposed method in practice.