How to Interpret AHP/ANP Application Results in a Really Meaningful Manner?

Grzegorz Ginda

doi:10.7494/dmms.2021.15.7592

Authors

Grzegorz Ginda AGH University of Krakow, Krakow, Poland https://orcid.org/0000-0003-2153-1858

DOI:

https://doi.org/10.7494/dmms.2021.15.7592

Keywords:

decision analysis, AHP/ANP, results, interpretation, compatibility, qualitative

Abstract

Final decision recommendations rely heavily on ranking Decision-Making Units (DMUs), often achieved using Saaty’s Analytic Hierarchy/Network Process (AHP/ANP). AHP/ANP provides precise overall priority scores which decision-makers commonly treat as definitive for ranking purposes. This reliance means that even minimal numerical differences between DMUs are used to determine the final selection. However, this strict adherence to tiny numerical distinctions – disregarding the actual degree of difference – is problematic. Practically, it risks rejecting DMUs whose performance is only slightly inferior; methodologically, it contradicts the qualitative nature of the input (pairwise comparisons) with the quantitative output. This tension raises the question of achieving an adequate qualitative interpretation of the quantitative rankings. To resolve this, the paper proposes clustering approaches to help decision-makers reliably group and discriminate among similar DMUs. These methods aim to justify more informed choices by avoiding spurious precision. The approaches were tested using two diverse decision cases. The results are promising and indicate that these clustering techniques can be useful under certain specific circumstances.

References

Bezdek J.C. & Pal N.R. (1998). Some new indexes of cluster validity. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 28(3), pp. 301–315. DOI: https://doi.org/10.1109/3477.678624.

Dunn J.C. (1973). A fuzzy relative of the ISODATA process and its use in detecting compact well-separated clusters. Journal of Cybernetics, 3(3), pp. 32–57. DOI: https://doi.org/10.1080/01969727308546046.

Florek K., Łukaszewicz J., Perkal J., Steinhaus H. & Zubrzycki S. (1951). Taksonomia numeryczna wrocławska. Przegląd Antropologiczny, 17, pp. 193–211.

Ginda G. & Szyba M. (2020). How different are Polish regions with regard to the utilisation of biogas plant technology application potential? [In:] ISAHP 2020 Web conference. URL: https://isahp2020.exordo.com/files/papers/107/final_draft/ggms_isahp2020.pdf [10.12.2020].

Ishizaka A., Nemery P. & Pearman C. (2012). AHPSort: An AHP based method for sorting problems. International Journal of Production Research, 50(17), pp. 4767–4784. DOI: https://doi.org/10.1080/00207543.2012.657966.

Ismkhan H. (2017). A.1D-C: A novel fast automatic heuristic to handle large-scale one-dimensional clustering. Applied Soft Computing, 52, pp. 4767–4784. DOI: https://doi.org/10.1016/j.asoc.2016.09.001.

Kaufman L. & Rousseeuw P.J. (1990). Finding Groups in Data: An Introduction to Cluster Analysis. Hoboken, NJ: Wiley-Interscience. DOI: https://doi.org/10.1002/9780470316801.

Kolenda M. (2006). Taksonomia numeryczna. Klasyfikacja, porządkowanie i analiza obiektów wielocechowych. Wrocław: Wydawnictwo Akademii Ekonomicznej im. Oskara Langego.

Mankowski M.A. & Moshkov M.J. (2021). One-dimensional clustering. Studies in Systems, Decision and Control, 331, pp. 123–133.

Netto A.L., Salomon V.A.P., Ortiz-Barros M.A., Florek-Paszkowska A.K., Petrillo A. & De Oliveira O. (2021). Multiple criteria assessment of sustainability programs in the textile industry. International Transactions in Operational Research, 28, pp. 1550–1572. DOI: https://doi.org/10.1111/itor.12871.

Opricovic S. (1988). Multicriteria Optimization of Civil Engineering Systems. Belgrade: Faculty of Civil Engineering, University of Belgrade.

Opricovic S. & Tzeng G.-H. (2004). Compromise solution by MCDM methods: A comparative analysis of VIKOR and TOPSIS. European Journal of Operational Research, 156(2), pp. 445–455. DOI: https://doi.org/10.1016/S0377-2217(03)00020-1.

Pakhira M.K., Bandyopadhyay S. & Maulik U. (2004). Validity index for crisp and fuzzy clusters. Pattern Recognition 37(3), pp. 487–501. DOI: https://doi.org/10.1016/j.patcog.2003.06.005.

Saaty T.L. (1980). The Analytic Hierarchy Process: Planning, Priority Setting, Resource Allocation, New York: McGraw-Hill.

Saaty T.L. (1996). Decision Making with Dependence and Feedback. Pittsburgh: RWS.

Spaeth H. (1973). Algorithm of clustering of one-dimensional data ordered data. Computing, 11(2), pp. 175–177. DOI: https://doi.org/10.1007/BF02252907.

Szczotka F.A. (1972). On the method of ordering and clustering objects. Zastosowania Matematyki, XIII(1), pp. 23–34.

Szkaliczki T. (2016). Clustering.sc.dp: Optimal clustering with sequential constraint by using dynamic programming. R Journal, 8(1), pp. 318–327. DOI: https://doi.org/10.32614/rj-2016-022.

Tibshirani R., Walter G. & Hastie T. (2001). Estimating the number of clusters in a data set via the gap statistic. Journal of the Royal Statistical Society. Series B: Statistical Methodology, 63(2), pp. 411–423. DOI: https://doi.org/10.1111/1467-9868.00293.

Wang H. & Song M. (2011). Ckmeans.1d.dp: Optimal k-means clustering in one dimension by dynamic programming. R Journal 3(2), pp. 29–33. DOI: https://doi.org/10.32614/RJ-2011-015.