Time-critical interactive dynamic influence diagram

Multiagent time-critical dynamic decision making is a challenging task in many real-world applications where a trade-off between solution quality and computational tractability is required. In this paper, we present a formal representation for modelling time-critical multiagent dynamic decision problems based on interactive dynamic influence diagrams (I-DIDs). The new representation called time-critical I-DIDs (TC-IDIDs) represents space-temporal abstraction by providing time-index to nodes and the model is defined in terms of the condensed and deployed forms. The condensed form is a static model of TC-IDIDs and can be expanded into its dynamic version. To facilitate the conversion between the two forms, we exploit the notion of object-orientation design to develop flexible and reusable TC-IDIDs. The difficulty on expanding TC-IDIDs is to select a proper time sequence to index nodes in the condensed form so that the expanded TC-IDIDs can be solved efficiently without compromising the quality of the policy. For this purpose, we propose two methods to build the condensed form of TC-IDIDs. We evaluate the solution quality and time complexity in three well-studied problems and provide results in support.     

A note on system modelling and control in the case of conflicting objectives

The paper is concerned with the problem of system modelling in the case of many decision makers with conflicting goals. The suggested approach, preserving much of the tractability of the standard LQG problems, allows to take into account decision makers' risk attitudes. This revised version was published online in June 2006 with corrections to the Cover Date.     

Genetic learning of accurate and compact fuzzy rule based systems based on the 2-tuples linguistic representation

One of the problems that focus the research in the linguistic fuzzy modeling area is the trade-off between interpretability and accuracy. To deal with this problem, different approaches can be found in the literature. Recently, a new linguistic rule representation model was presented to perform a genetic lateral tuning of membership functions. It is based on the linguistic 2-tuples representation that allows the lateral displacement of a label considering an unique parameter. This way to work involves a reduction of the search space that eases the derivation of optimal models and therefore, improves the mentioned trade-off.Based on the 2-tuples rule representation, this work proposes a new method to obtain linguistic fuzzy systems by means of an evolutionary learning of the data base a priori (number of labels and lateral displacements) and a simple rule generation method to quickly learn the associated rule base. Since this rule generation method is run from each data base definition generated by the evolutionary algorithm, its selection is an important aspect. In this work, we also propose two new ad hoc data-driven rule generation methods, analyzing the influence of them and other rule generation methods in the proposed learning approach. The developed algorithms will be tested considering two different real-world problems.     

Stability analysis of data-driven local model networks

This article discusses stability analysis of data-driven dynamic local model networks. In contrast to traditional fuzzy modelling, the structure and complexity of such model architectures is not unique when only observed input- and output data are available for their parametrization. The present article complements the well-known trade-off between accuracy and complexity by the notion of stability. For this purpose, existing Lyapunov stability criteria for local model networks are extended by a decay rate which represents a scalar and quantitative stability measure. It allows to compare models with different degrees of complexity also in view of their stability. For some of the commonly available Lyapunov stability criteria, the individual local model transitions are crucial. Therefore, in this article, an approach is introduced to determine the actually occurring model transitions by means of the identification data. The methods presented in the article are illustrated and discussed by means of a simulation example. It is shown how model complexity and the related approximation quality can have an adverse impact on the stability and how the outcome of different Lyapunov criteria is affected by the proper determination of local model transitions.     

Dimension reduction in multiobjective optimization

In multiobjective optimization there is often the problem of the existence of a large number of objectives. For more than two objectives there is a difficulty with the representation and visualization of the solutions in the objective space. Therefore, it is not clear for the decision maker the trade-off between the different alternative solutions. Thus, this creates enormous difficulties when choosing a solution from the Pareto-optimal set and constitutes a central question in the process of decision making. Based on statistical methods as Principle Component Analysis and Cluster Analysis, the problem of reduction of the number of objectives is addressed. Several test examples with different number of objectives have been studied in order to evaluate the process of decision making through these methods. Preliminary results indicate that this statistical approach can be a valuable tool on decision making in multiobjective optimization. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Rationalisation of building maintenance by Markov decision models: a pilot case study

In this paper we develop a Markov decision model for rationalising building maintenance at a strategic level. Through a global modelling of deterioration we are able to determine the maintenance policy that ensures a specified average quality level at minimal cost. In this way we are able to produce a trade-off curve between overall quality level vs the minimum required level of maintenance costs. The model is applied in a pilot case for four main building elements, viz. masonry, pointing, window frames and painting of buildings owned by a building society. This demonstrates the value of the model as management instrument to determine and allocate budgets.     

The transformed inverse Gaussian process as an age- and state-dependent degradation model

In this paper, a transformed inverse Gaussian (TIG) process is introduced as a new family of monotonic degradation models. Different from most state-of-the-art degradation models, which can only characterize age-dependent performance degradation, the TIG process model is mainly introduced for degradation modelling of industrial products with age- and state-dependent performance degradation. With this new model, promising properties include (1) the modelling capability for characterizing products observed at discrete time points with age- and state-dependent degradation, (2) the mathematical tractability for calculating the reliability function and remaining useful life distribution with high efficiency, and (3) the modelling flexibility of incorporating explanatory variables and random effects for investigating a product population with unit-to-unit heterogeneity. To facilitate the degradation modelling and analysis, methods for parameter estimation and model selection are developed under a coherent Bayesian framework. Simulation studies and real cases are presented to demonstrate the proposed degradation model and the Bayesian methods.     

Knowledge Representation, Reasoning and Integration Using Temporal Logic with Clocks

Representation, reasoning about and integrating knowledge based on multiple time granularities in knowledge-based systems is important, especially when talking about events that take place in the real world. Formal approaches based on temporal logics have been successfully applied in many application domains of knowledge-based systems where the evolution of a system and its environment through time is central. This paper presents a methodology based on temporal logic to deal with knowledge based on multiple time granularities in knowledge-based systems. The temporal logic we consider is especially suitable for modelling events with different rates and/or scales of progress. The methodology includes an approach to the representation of timing systems, a method used for representing facts and rules in a knowledge-based system that involve multiple time granularities using temporal logic, and several deductive reasoning techniques. The work presented in this article has been supported in part by The Australian Research Council and Macquarie University. Note that this paper is an extended and revised version of Orgun, Liu and Nayak [37].     

Expert systems approach to decision support

An approach to building decision support systems based on expert systems methods is proposed. The relatively weak basic assumptions include only stationarity (repeatability) of the decision in the same environment (circumstances) and ordering of the values of all the attributes with regard to the decision maker's preferences. The proposed approach is aimed at reflecting the experienced domain expert's and decision maker's knowledge and preferences, both in the form of facts and rasoning rules. Among the purposes of the described class of decision support systems there are the storage and retrieval of the expert's knowledge and decisions, decision making support and ranking of admissible decision alternatives. A general model of the decision process is proposed and a language for representation of the expert's knowledge is introduced in brief. The structure, reasoning control, and an example of application of the proposed system are discussed and possible further extensions are pointed out.     

Multicriteria Analysis Represented by Artificial Intelligence Techniques

This paper presents a possible representation of multicriteria analysis by means of artificial intelligence techniques. The decision process activities characterized by the existence of formal and technical knowledge were identified and attention was focused on the area of multicriteria outranking methods. The knowledge characteristics suggested the use of artificial intelligence techniques, based on a conceptualization in which the domain of discourse is the set of the multicriteria methodology concepts used in the analysed area of activities, and the relational set is the union of the admissible relations among the concepts and the relations elicited from experience. The suitable AI techniques were tested by implementing a knowledge-based interface between the outranking methods and a user who was not very familiar with this approach.     

Multiobjective strategies in power systems planning

The rapidly changing decision environment in electrical systems planning calls for a new approach in planning procedures and mathematical tools. The method proposed in this study decomposes the decision making into two strata, one at the strategy setting, the other at the project selection level. For the strategy setting level, a multiobjective dynamic linear programming model is used in generating system expansion decision alternatives. The suitability of the modelling approach is considered from the various aspects such as congruence, flexibility, transparency, and accuracy.An application of the model to the Turkish electrical system is also presented, where expansion alternatives to the year 2000 are generated. Three criteria were observed: cost, environmental impact and risk.     

Representation of decision making processes in an ecological evaluation system

This paper deals with the problems of knowledge representation for a decision support system (DSS) applicable in a dynamic environment. Some special principles concerning environment applications are considered in order to understand better human decision making and behavior. The approach of representing static and dynamic aspects of a system and reflecting them using deep knowledge representation is proposed. The formalization of multiple objective decision making mechanisms is considered. The results of modeling cognitive processes leading to decisions are demonstrated by an example developed during the design stages of an ecological evaluation system.     

Structure and logic in knowledge-based representations of systems simulation models

The paper focuses on the similarity between modelling and knowledge representation, trying to bring together the OR/Systems Science and the Artificial Intelligence views when referring to a computer system simulation, especially of the discrete-event or the network types. The models we consider are generalized activity networks with resources, including either models with a finite lifetime, such as project scheduling networks, or steady state models, such as queueing networks. By enhancing the structure of entities and states and the logic of transitions within a model specification, modularity is improved and one may adopt a more declarative approach. The relational and rule-based representation formalisms are a convenient choice for that purpose. Then, the use of knowledge bases both for the static (i.e. consultative) and the dynamic (i.e. experimental) study of the model turns up to be more natural. Moreover, the task of building an expert system for decision support on system analysis or synthesis becomes easier. The paper reports some original work in the above directions, using a logic programming approach and an associated specification methodology based on general systems concepts.     

Fast computation of bounds for two-terminal network reliability

In this paper, an algorithm for the fast computation of network reliability bounds is proposed. The evaluation of the network reliability is an intractable problem for very large networks, and hence approximate solutions based on reliability bounds have assumed importance. The proposed bounds computation algorithm is based on an efficient BDD representation of the reliability graph model and a novel search technique to find important minpaths/mincuts to quickly reduce the gap between the reliability upper and lower bounds. Furthermore, our algorithm allows the control of the gap between the two bounds by controlling the overall execution time. Therefore, a trade-off between prediction accuracy and computational resources can be easily made in our approach. The numerical results are presented for large real example reliability graphs to show the efficacy of our approach.     

Robust optimization of uncertain multistage inventory systems with inexact data in decision rules

In production-inventory problems customer demand is often subject to uncertainty. Therefore, it is challenging to design production plans that satisfy both demand and a set of constraints on e.g. production capacity and required inventory levels. Adjustable robust optimization (ARO) is a technique to solve these dynamic (multistage) production-inventory problems. In ARO, the decision in each stage is a function of the data on the realizations of the uncertain demand gathered from the previous periods. These data, however, are often inaccurate; there is much evidence in the information management literature that data quality in inventory systems is often poor. Reliance on data “as is” may then lead to poor performance of “data-driven” methods such as ARO. In this paper, we remedy this weakness of ARO by introducing a model that treats past data itself as an uncertain model parameter. We show that computational tractability of the robust counterparts associated with this extension of ARO is still maintained. The benefits of the new model are demonstrated by a numerical test case of a well-studied production-inventory problem. Our approach is also applicable to other ARO models outside the realm of production-inventory planning.     

A disjunctive graph model and framework for constructing new train schedules

Train scheduling is a complex and time consuming task of vital importance in many countries. To create completely new train schedules that are more accurate and efficient than permitted by current techniques, a novel “hybrid” job shop approach is proposed and implemented in this paper. Unique characteristics of train scheduling are firstly incorporated into a disjunctive graph representation of the solution. Dedicated “stand-alone” constructive algorithms that utilise this representation are then developed. The modelling approach and the constructive algorithms are essential as they provide the basis for which meta-heuristics and other iterative refinement algorithms can be applied. A numerical investigation and case study is provided and demonstrates the viability of the modelling approach. Furthermore it is demonstrated that good quality solutions are provided with reasonable computational effort.     

An adaptive TS approach to JIT sequencing with variable processing times and sequence-dependent setups

This paper addresses a single machine sequencing problem with variable processing times and sequence-dependent setups. The objective is to find the best trade-off between the JIT goal and the processing time compression and extension costs by simultaneously determining the job sequence and processing times for concerned jobs. Due to the combinatorial nature of the problem, it cannot be optimally solved in polynomial time. A tabu search approach is used to provide good and quick solutions. To improve the computational efficiency, an adaptive neighbourhood generation method is proposed and used in the tabu search algorithm. A total of 100 problems of different sizes have been solved to test the proposed approach. Our computational experience shows that the adaptive approach outperforms several other neighbourhood generation methods in terms of both convergence rate and solution quality. The effects of the search parameters are also discussed.     

Multicriteria evidential reasoning decision modelling and analysis—prioritizing voices of customer

In this paper, a new methodology is investigated to support the prioritization of the voices of customers through various customer satisfaction surveys. This new methodology consists of two key components: an innovative evidence-driven decision modelling framework for representing and transforming large amounts of data sets and a generic reasoning-based decision support process for aggregating evidence to prioritize the voices of customer on the basis of the Evidential Reasoning (ER) approach. Methods and frameworks for data collection and representation via multiple customer satisfaction surveys were examined first and the distinctive features of quantitative and qualitative survey data are analysed. Several novel yet natural and pragmatic rule-based functions are then proposed to transform survey data systematically and consistently from different measurement scales to a common scale, with the original features and profiles of the data preserved in the transformation process. These new transformation functions are proposed to mimic expert judgement processes and designed to be sufficiently flexible and rigorous so that expert judgements and domain specific knowledge can be taken into account naturally, systematically and consistently in the transformation process. The ER approach is used for synthesizing quantitative and qualitative data under uncertainty that can be caused due to missing data and ambiguous survey questions. A new generic method is also proposed for ranking the voices of customer based on qualitative measurement scales without having to quantify assessment grades to fixed numerical values. A case study is examined using an Intelligent Decision System (IDS) to illustrate the application of the decision modelling framework and decision support process for prioritizing the voices of customer for a world-leading car manufacturer.