This paper describes and evaluates approaches to model and solve technical product configuration problems using different artificial intelligence methodologies. By means of a typical example, the benefits and limitations of different artificial intelligence methods are discussed and a flexible software architecture for integrating different solvers in a product configurator is proposed.

Product configuration is a key technology, which enables businesses to deliver and deploy individualized products. In many cases, finding the optimal configuration solution for the user is a creative process that requires them to decide trade-offs between conflicting goals (multicriteria optimization problem). These problems are best supported by an interactive dialog that is managed by a dedicated software program (the configurator) that provides decision support. We illustrate this using a real example (configuration of a business software system). This productively used application makes the user aware of which choices are available in a given situation, provides assistance in resolving inconsistent choices and defaults, and generates explanations if desired. One of the key configurator components used to manage this is a truth maintenance system. We describe how this component is used and two novel extensions to it: methods for declarative handling of defaults (of varying strength) and the declarative handling of incompleteness. Finally, we summarize our experiences made during the implementation of this application and the pros and cons of declarative versus procedural approaches.

This article describes the technical principles and representation of a constraint-based configuration method for work processes. Methods developed for the configuration of modular systems comprising components have traditionally adopted a representation where the properties and compatibility requirements are expressed as constraints associated with individual components. However, this representation does not accurately capture constraints on paths and subprocesses and is therefore unsuitable for process configuration. This article extends established constraint-based configuration methods with a constraint language for specifying properties of execution paths in work processes. A framework for semiautomated process customization is presented. It integrates the extended constraint language with a metamodel of the work processes in an organization and allows to adapt generic work processes to fit the requirements of specific development projects. Heuristic search methods are applied to build valid process configurations by incrementally resolving constraint violations. The declarative framework facilitates the adaptation of abstract work processes as well as the validation and repair of existing processes. The approach was developed in the context of a real-world system of complex design and development processes where it was shown that significant process improvements and reduction in effort required to edit process models can be achieved.

Product configuration is a major industrial application domain for constraint satisfaction techniques. Conditional constraint satisfaction problems (CCSPs) and feature models (FMs) have been developed to represent configuration problems in a natural way. CCSPs are like constraint satisfaction problems (CSPs), but they also include potential variables, which might or might not exist in any given solution, as well as classical variables, which are required to take a value in every solution. CCSPs model, for example, options on a car, for which the style of sunroof (a variable) only makes sense if the car has a sunroof at all. FMs are directed acyclic graphs of features with constraints on edges. FMs model, for example, cell phone features, where utility functions are required, but the particular utility function “games” is optional, but requires Java support. We show that existing techniques from formal methods and answer set programming can be used to naturally model CCSPs and FMs. We demonstrate configurators in both approaches. An advantage of these approaches is that the model builder does not have to reformulate the CCSP or FM into a classic CSP, converting potential variables into classical variables by adding a “does not exist” value and modifying the problem constraints. Our configurators automatically reason about the model itself, enumerating all solutions and discovering several kinds of model flaws.

Knowledge-based configurators are supporting configuration tasks for complex products such as telecommunication systems, computers, or financial services. Product configurations have to fulfill the requirements articulated by the user and the constraints contained in the configuration knowledge base. If the user requirements are inconsistent with the constraints in the configuration knowledge base, users have to be supported in finding out a way from the no solution could be found dilemma. In this paper we introduce a new algorithm (PersDiag) that allows the determination of personalized diagnoses for inconsistent user requirements in knowledge-based configuration scenarios. We present the results of an empirical study that show the advantages of our approach in terms of prediction quality and efficiency.

Configurators have been generally accepted as important tools to elicit customers' needs and find the matches between customers' requirements and company's offerings. With product configurators, product design is reduced to a series of selections of attribute values. However, it has been acknowledged that customers are not patient enough to configure a long list of attributes. Therefore, making every round of configuring process productive and hence reducing the number of inputs from customers are of substantial interest to academic and industry alike. In this paper, we present an efficient product configuration approach by incorporating Shapley value, which is a concept used in game theory, to estimate the usefulness of each attribute in the configurator design. This new method iteratively selects the most relevant attribute that can contribute most in terms of information content from the remaining pool of unspecified attributes. As a result from product providers' perspective, each round of configuration can best narrow down the choices with given amount of time. The selection of the next round query is based on the customer's decision on the previous rounds. The interactive process thus runs in an adaptive manner that different customers will have different query sequences. The probability ranking principle is also exploited to give product recommendation to truncate the configuration process so that customers will not be burdened with trivial selection of attributes. Analytical results and numerical examples are also used to exemplify and demonstrate the viability of the method.

This paper presents a study of how the use of product configurators affects business processes of engineering-oriented companies. A literature study shows that only a minor part of product configuration research deals with the effects of product configuration, and that the ones that do are mostly vague when reporting the effects of configurator projects. Only six cases were identified, which provide estimates of the actual size of lead time reduction achieved from product configurators. To broaden this knowledge, this paper presents the results of a study of 14 companies concerning the impact of product configurators on business processes related to the creation of quotes and detailed product specifications. The study documents impressive results of the application of configurator technology. For example, in the data retrieved the use of configurators was estimated to have implied up to a 99.9% reduction of the quotation lead time with an average estimated reduction of 85.5%.