Support for configuration of physical products and services

Juha Tiihonen

Forskningsoutput: AvhandlingMagisteruppsatsAvhandlingar


The ideal of mass customization is to satisfy individual customer requirements efficiently. This can be realized with configurable products that can be adapted, within the bounds of pre-designed offered variation, to individual requirements. Configurators are information systems that support efficient and errorless specification of individualized products or services. Applying the Design Science research approach, a number of artifacts that support the sales configuration of physical products and services were developed with the aim of advancing the state of the art of practically applicable configurators. We present a conceptualization for configuration knowledge that unifies previous connection-based, resource-based, structure-based, and function-based approaches. The conceptualization is object oriented and treats the main concepts uniformly with respect to several criteria, including the availability of taxonomic hierarchies with refinement, abstraction, and applicability of attributes. A detailed conceptualization for representing the variable compositional structure of components and functions is included. The main artifact of this work is a novel configurator instantiation called WeCoTin. It consists of a graphical modeling environment Modeling Tool and a Web-based Configuration Tool that supports the configuration task. WeCoTin is based on a well-founded modeling conceptualization and a corresponding high-level object-oriented modeling language with clear formal semantics, provided by mapping the modeling language to weight constraint rules—a form of logic programs. The Modeling Tool enables efficient graphical modeling and includes features that support long-term management. The Configuration Tool provides Web-based sales configuration functionality. It applies an inference engine that is based on weight constraint rules to provide consistent and complete inference. Evaluation includes the characterization of 26 sales configuration models and run-time performance analysis. Enabled by the novel principles of WeCoTin, an information systems design theory for sales configurators is proposed. Recommendation technologies can support users during choice navigation. We present scenarios in which support can be offered and analyze the applicability of recommendation technologies. We propose extensions to the existing case-based feature value recommendation technologies by integrating importance weights and similarity metrics. A basic evaluation of the utility of the case-based collaborative approach was provided through an empirical study. Due to the importance of services, mass customization of services by configuration is crucial. We discuss the offered variation of configurable services in three industries and the applicability of configurators designed for physical products in the context of service configuration.
Tryckta ISBN978-952-60-5894-8
StatusPublicerad - 2014
MoE-publikationstypG2 Masteruppsats, polyteknisk masteruppsats

Tilläggsuppgifter om doktorsavhandling

PhD dissertation in reality, the system only allows M.Sc.

Citera det här