|
BPMSG |
 |
|
Goepel, K. D. 2018, July 12—15 Implementation of an online software tool for the analytic hierarchy process (AHP-OS). Paper presented at International Symposium of the Analytic Hierarchy Process (ISAHP 2018). Added by: Klaus D. Goepel (08 Jun 2019 07:36:46 Asia/Singapore) Last edited by: Klaus D. Goepel (08 Jun 2019 07:38:11 Asia/Singapore) |
| Resource type: Proceedings Article DOI: 10.13033/isahp.y2018.029 BibTeX citation key: Goepel2018a Email resource to friend View all bibliographic details  |
Categories: AHP/ANP, Decision Making Subcategories: Decision making support systems Keywords: AHP online, AHP software, AHP-OS Creators: Goepel Collection: International Symposium of the Analytic Hierarchy Process (ISAHP 2018) |
Views: 9/540 Views index: % Popularity index: 12% |
| Abstract |
|
The analytic hierarchy process (AHP) remains a popular multi-criteria decision method. The author has implemented a free, web-based AHP online system with noteworthy features, allowing for the detailed analysis of decision problems. Beside standard functions like flexible decision hierarchies, support to improve inconsistent judgments, alternative evaluation and sensitivity analysis, the software can handle group inputs, calculate group consensus based on Shannon α and β-entropy, and estimate weight uncertainties based on randomized small variations of input judgments. In addition, different AHP judgment scales can be applied a posteriori, and alternative evaluation can be done using the weighted sum (WSM) or weighted product model (WPM). This flexibility opens up opportunities to study the classical AHP and decision projects under various parameters. The author’s intention was to provide a complete and free software tool for educational and research purposes, where calculations and algorithms are well documented and all input data and results can be exported in an open format for further processing or presentation. The article describes the basic structure of the software and highlights key features of its implementation. The full description of all underlying methods and algorithms is available from the author’s website.
|