Predictive analytics and disused railways requalification: Insights from a Post Factum Analysis perspective

Abstract

Strategic decision making problems in the public policy domain typically involve the comparison of competing options by different stakeholders. This paper considers a real case study oriented toward ranking potential actions for the regeneration of disused railways in Italy. The study involves multiple conflicting criteria such as an expected duration of construction works, costs, a number of potential users, and new green areas. Within this context, we demonstrate that Post Factum Analysis (PFA) coupled with Decision Aiding supports the development of robust recommendations. The role of PFA is to highlight how the actions' performances need to be modified so that the recommendation is changed in a desired way. In particular, it highlights the minimal improvements that would warrant the feasibility of some currently impossible outcome (e.g., achieving a better position in the ranking) or the maximal deteriorations that alternatives can afford to maintain some target result (e.g., not losing their advantage over some other options). The use of a focus group with both experts and participants in the decision making process provided insights on how PFA can support: (i) the creation of arguments in favour or against the respective options under analysis, (ii) understanding of the results' sensitivity with respect to possible changes in the alternatives' performances, (iii) a better informed discussion about the results among the participants in the process, and (iv) the development of new/better alternatives.

How to support the application of multiple criteria decision analysis? Let us start with a comprehensive taxonomy

“…Another solution to test results legitimacy is to exploit sensitivity analysis, which can be used to study if and how the recommendation already derived with a particular method would be affected under different alternatives performance and/or DM’s preferences. Examples in this regard are the works by Ferretti et al [ 54 ] and Ciomek et al [ 33 ], where sensitivity on weighting and performances, respectively, is used to evaluate the stability of the best alternatives, attained ranks, or pairwise preference relations.…”

“…The use of MCDA has increased consistently over the years, due to recognition of the need to address challenges from a multitude of perspectives while dealing with several trade-offs, which calls for multiple criteria assessments [ 24 , 94 , 51 , 73 ]. This trend has been particularly marked in complex decision making domains like energy technologies and systems evaluations [ 169 ], urban regeneration planning [ 33 ], policy analysis [ 138 ], health technology management [ 131 ], ecosystem services governance [ 109 ], sensors placement [ 13 ], resilience and sustainability quantification [ 107 , 61 ], energy policy ranking [ 42 ], product recovery activities [ 9 ], supply chain management [ 67 ], waste recycling [ 66 ], freight selector [ 53 ], housing affordability [ 122 ], raw material supply risk assessment [ 87 ], water supply systems [ 149 ], and polluted land remediation [ 81 ].…”

Reuse of Abandoned Railways Leads to Urban Regeneration: A Tale from a Rust Track to a Green Corridor in Zhangjiakou

“…Learning the multiattribute value theory (SAVT), they evaluated the psychological goals of comprehensive stakeholders and offered measurable quantitative assessment schemes to determine the best ways of transforming the abandoned railway. A post-event analysis(PFA) model [29], an analytic network process (ANP) method [30], zero-one goal programming (ZOGP) [31] and an indicator system for sustainability [32] have been used for a comprehensive assessment to determine the optimal reuse solution of abandoned railways.…”

Long-term multi-criteria improvement planning

“…Different families of MCDA methods have been proposed in the literature and vary depending on the underlying assumptions related to the preference structure of the decision makers. These have been applied in different fields (Behzadian et al, 2010), including budgeting and resource allocation (Lootsma et al, 1990), geographical information systems (Ligmann-Zielinska & Jankowski, 2012), Another recent contribution that involves multi-criteria improvement planning is Post Factum Analysis (PFA) proposed by Kadziński et al (2016) and later applied and developed by Ciomek et al (2018), as well as Dutta et al (2019). It is strongly related to the robustness concern associated with a given recommendation.…”