Abstract

Information systems based on Multi-Criteria Decision Analysis (MCDA) methods enable considering multiple attributes with contrary objectives. Information systems using MCDA simplify and automatize assessment toward automatizing decision support systems. Individual MCDA methods differ in their algorithms, implying different results for the same problem. Moreover, the diversity of algorithms refers to the MCDA methods and their techniques used at an individual stage, such as distance metrics. They are implemented in MCDA methods to measure alternatives’ distances from reference solutions. The most commonly used metric is the Euclidean distance. However, other distance metrics are also suitable for this purpose. Moreover, a broad set of metrics can be helpful in comparative analysis to test the robustness of particular scenarios. Therefore, the main contribution of a Python library for multi-criteria decision analysis called distance-metrics-mcda is providing a set of 20 distance metrics for benchmarking purposes. The implemented library offers an autonomous tool for evaluating any decision problem. The presented library is an important addition to decision support systems based on MCDA methods as it provides additional possibilities for analysis of scenarios’ reliability.

Recommended Citation

Baczkiewicz, A., Watróbski, J., & Sałabun, W. (2022). Distance Metrics Library for MCDA Methods. In R. A. Buchmann, G. C. Silaghi, D. Bufnea, V. Niculescu, G. Czibula, C. Barry, M. Lang, H. Linger, & C. Schneider (Eds.), Information Systems Development: Artificial Intelligence for Information Systems Development and Operations (ISD2022 Proceedings). Cluj-Napoca, Romania: Babeș-Bolyai University.

Share

COinS
 

Distance Metrics Library for MCDA Methods

Information systems based on Multi-Criteria Decision Analysis (MCDA) methods enable considering multiple attributes with contrary objectives. Information systems using MCDA simplify and automatize assessment toward automatizing decision support systems. Individual MCDA methods differ in their algorithms, implying different results for the same problem. Moreover, the diversity of algorithms refers to the MCDA methods and their techniques used at an individual stage, such as distance metrics. They are implemented in MCDA methods to measure alternatives’ distances from reference solutions. The most commonly used metric is the Euclidean distance. However, other distance metrics are also suitable for this purpose. Moreover, a broad set of metrics can be helpful in comparative analysis to test the robustness of particular scenarios. Therefore, the main contribution of a Python library for multi-criteria decision analysis called distance-metrics-mcda is providing a set of 20 distance metrics for benchmarking purposes. The implemented library offers an autonomous tool for evaluating any decision problem. The presented library is an important addition to decision support systems based on MCDA methods as it provides additional possibilities for analysis of scenarios’ reliability.