Solving influence diagrams: efficient mixed-integer programming formulation and heuristic

Helmi Hankimaa, Olli Herrala, Fabricio Oliveira, Jaan Tollander de Balsch

Confirmation: I have read and agree with the workshop's policy on behalf of myself and my co-authors.

Tracks: Main Track

Keywords: decision problems under uncertainty, influence diagrams, decision analysis, mixed-integer programming

TL;DR: We introduce novel MIP formulations and a heuristic for solving influence diagrams efficiently, achieving significant computational gains and demonstrating their applicability in a real-world preventive healthcare case study.

Abstract: We propose novel mixed-integer linear programming (MIP) formulations to model decision problems posed as influence diagrams. We also present a novel heuristic that can be employed to warm start the MIP solver and provide heuristic solutions to more computationally challenging problems. We provide computational results showcasing the superior performance of these improved formulations as well as the performance of the proposed heuristic. Lastly, we describe a novel case study showcasing decision programming as an alternative framework for modelling multi-stage stochastic dynamic programming problems.

Submission Number: 31