(343c) Reconciling Ranking Criteria with Fuzzy Sets for Effective Use in Project Portfolio Selection

Authors: 
Bury, S. J., Dow Inc.
Amaran, S., Dow Inc.
Rajagopalan, S., Dow Inc.
Agarwal, A., Carnegie Mellon University
Maintenance activities at process plants can be divided into “run-the-plant” and improvement categories. The first category is focused on the daily maintenance activities that ensure safe plant operation. The maintenance improvement category are larger projects aimed at further increasing operating reliability. Within a specified investment budget, it is a challenge to determine the right mix of these improvement projects that will meet performance and economic targets. One approach is portfolio optimization. Although optimization for Project Portfolio selection has been researched since the 1950’s resulting in models with high degrees of mathematical sophistication (Salo et al 2011), there is a significant gap between the extents of application in the financial sector compared with project investment selection, for example, the selection of maintenance improvement projects at multiple process plants within a specified budget. A contributing factor to the lack of application is that the selection of maintenance improvement projects is complicated by scaling issues of consequences and decision criteria that are different than traditional monetary considerations like an expected rate of return. A recent paper discusses this complexity for portfolio selection of maintenance infrastructure improvement projects (Mild 2015). One common approach for multiple quantitative and qualitative criteria is to use 1,3,9 ranking matrices. It can be a challenge to harmonize these rankings especially when they are provided by a large cohort of SME’s across disparate plants. Ayyub suggested that for system variables that are qualitative in nature, fuzzy set theory can be used in defining the potential states together with a suitable observation channel that yields a quantitative equivalent for each state (Ayyub 2001). In this work we demonstrate how we implemented this idea for the selection of an optimal portfolio of maintenance improvement projects from a list of project proposals from 1300 plants using fuzzy set theory based on the work done by Dubois and Prade (Dubois 1987), Carlsson et al. (Carlsson 2001), and Vercher et al.(Vercher 2007).

Salo, J. Keisler and A. Morton (eds.) Advances in Portfolio Decision Analysis: Improved Methods for Resource Allocation. 2011

Mild, Juuso Liesiö, Ahti Salo, Selecting infrastructure maintenance projects with Robust Portfolio Modeling, Decision Support Systems, Volume 77, 2015

Ayyub, Bilal M. Elicitation of expert opinions for uncertainty and risks 2001

Dubois and H. Prade. The mean value of a fuzzy number. Fuzzy Sets and Systems, 24:279-300, 1987.

Carrlson, Fuller, R. and Majlender, P. A possibilistic approach to selecting portfolios with highest utility score. TUCS Technical Report No. 355, August 2000