Peter Keenan

Spatial decision support systems for vehicle routing

Abstract The vehicle routing field is a well-developed area of management science application. There is increasing recognition that effective decision-making in this field requires the incorporation of vehicle routing techniques into a decision support system (DSS). In order to provide decision support for a wide range of problems, routing techniques should be combined with systems that can take advantage of new technologies. These include spatial techniques drawn from the field of geographic information systems (GIS). A synthesis of appropriate algorithms and a GIS based computer system is identified as being necessary for effective decision support for the vehicle routing problem.

A hybrid genetic model for the prediction of corporate failure

Abstract.This study examines the potential of a neural network (NN) model, whose inputs and structure are automatically selected by means of a genetic algorithm (GA), for the prediction of corporate failure using information drawn from financial statements. The results of this model are compared with those of a linear discriminant analysis (LDA) model. Data from a matched sample of 178 publicly quoted, failed and non-failed, US firms, drawn from the period 1991 to 2000 is used to train and test the models. The best evolved neural network correctly classified 86.7 (76.6)% of the firms in the training set, one (three) year(s) prior to failure, and 80.7 (66.0)% in the out-of-sample validation set. The LDA model correctly categorised 81.7 (75.0)% and 76.0 (64.7)% respectively. The results provide support for a hypothesis that corporate failure can be anticipated, and that a hybrid GA/NN model can outperform an LDA model in this domain.

π Grammatical Evolution

πGrammatical Evolution is presented and its performance on four benchmark problems is reported. πGrammatical Evolution is a position-independent variation on Grammatical Evolution’s genotype-phenotype mapping process where the order of derivation sequence steps are no longer applied to nonterminals in a predefined fashion from left to right on the developing program. Instead the genome is used to specify which nonterminal will be developed next, in addition to specifying the rule that will be applied to that nonterminal. Results suggest that the adoption of a more flexible mapping process where the order of non-terminal expansion is not determined a-priori, but instead itself evolved, is beneficial for Grammatical Evolution.

Identifying online credit card fraud using Artificial Immune Systems

Significant payment flows now take place on-line, giving rise to a requirement for efficient and effective systems for the detection of credit card fraud. A particular aspect of this problem is that it is highly dynamic, as fraudsters continually adapt their strategies in response to the increasing sophistication of detection systems. Hence, system training by exposure to examples of previous examples of fraudulent transactions can lead to fraud detection systems which are susceptible to new patterns of fraudulent transactions. The nature of the problem suggests that Artificial Immune Systems (AIS) may have particular utility for inclusion in fraud detection systems as AIS can be constructed which can flag ‘non standard’ transactions without having seen examples of all possible such transactions during training of the algorithm. In this paper, we investigate the effectiveness of Artificial Immune Systems (AIS) for credit card fraud detection using a large dataset obtained from an on-line retailer. Three AIS algorithms were implemented and their performance was benchmarked against a logistic regression model. The results suggest that AIS algorithms have potential for inclusion in fraud detection systems but that further work is required to realize their full potential in this domain.

Modelling vehicle routing in GIS

The field of vehicle routing has seen the development of a variety of mathematical techniques. While the practical application of these techniques has depended on the use of spatial information such as road networks, these techniques have developed independently of other users of spatial information. This paper argues that traditional routing techniques have neglected the importance of path constraints and that GIS approaches allow the modelling of an extended range of routing problems. The well-known taxonomy of routing problems by Bodin and Golden is extended to reflect the additional possibilities introduced by the use of GIS techniques. The paper concludes by suggesting that the synthesis of vehicle routing and GIS techniques in a spatial decision support system can greatly enhance the modelling of these problems.

Cloud computing and DSS: the case of spatial DSS

While academic researchers in the field of decision support systems (DSS) tend to emphasise the technology independent nature of the decision support concept, they also need to assess the role of new technology; cloud computing is one such new technology. This paper suggests that the specific nature of DSS means that cloud computing is of limited relevance in most sectors of DSS application. However, spatial DSS is a distinct area of DSS application where large volumes of generic data are needed from outside the organisation making the decision. Consequently, a number of issues arise in the provision of data for SDSS which are not typical of the DSS field. Spatial data infrastructure (SDI) projects provide large collections of spatial data and can make use of technologies such as cloud computing. This paper argues that cloud computing can contribute to spatial DSS applications which use these large data resources. Spatial DSS remains a form of DSS which continues to push the limits of technology and developments in this sector can inform our understanding of the progression of the DSS field.

Grammar-Mediated Time-Series Prediction

Grammatical Evolution is a data-driven, model-induction tool, inspired by the biological gene-to-protein mapping process. This study examines the potential of Grammatical Evolution to uncover useful technical trading rulesets for intra-day equity trading. The form of these rule-sets is not specified ex-ante but emerges by means of an evolutionary process. High-frequency price data drawn from United States stock markets is used to train and test the model. The findings suggest that the developed rules earn positive returns in holdout test periods, and that the sizes of these returns are critically impacted by the choice of position exit-strategy.

Geographic Information and Analysis for Decision Support

Many types of decision making have a geographic (spatial) component. Geographic information systems (GIS) facilitate the organization and display of spatial data and provide a variety of distinctive spatial operations with that data. These functions allow decision makers explore the spatial aspects of their decisions. Consequently, GIS can be seen as an increasingly important technology for decision makers.

Arc routing for rural Irish networks

This paper discusses the capacitated arc routing problem in the context of large sparse networks drawn from sections of the Irish rural road network. The particular features of this form of arc routing are discussed. Various arc routing techniques are reviewed in this context. Two heuristics, a route first/cluster second heuristic and a clustering heuristic based on shortest path trees are proposed. These heuristics are shown to provide efficient routes and the clustering heuristic to provide routes which could form the basis of routes for practical use.

Solving large-scale time capacitated arc routing problems: from real-time heuristics to metaheuristics

This paper discusses the Time Capacitated Arc Routing Problem (TCARP) and introduces a heuristic and a metaheuristic algorithm for solving large-size instances of it. The TCARP is a realistic extension of the Capacitated Arc Routing Problem in which edge-servicing and edge-traversing costs, as well as vehicle capacities, are all time-based—i.e., given in time units. Accordingly, the TCARP goal is to minimise the total time employed in servicing the required edges, for which other edges might need to be traversed too. According to the numerical experiments carried out, the proposed heuristic is able to provide real-time results of high quality even for the largest instances considered. Likewise, the proposed metaheuristic outperforms other existing approaches, both in quality as well as in computing times.