Predicting Tactical Solutions to Operational Planning Problems under Imperfect Information

TL;DR

This paper introduces a machine learning methodology to rapidly predict tactical solutions for operational planning problems under imperfect information, enabling quick decision-making in complex, uncertain environments.

Contribution

It proposes a novel approach combining stochastic programming and supervised learning to efficiently generate accurate tactical solutions in operational planning.

Findings

Deep learning achieves high prediction accuracy

Predictions are generated in milliseconds or less

Performance comparable to traditional stochastic optimization methods

Abstract

This paper offers a methodological contribution at the intersection of machine learning and operations research. Namely, we propose a methodology to quickly predict tactical solutions to a given operational problem. In this context, the tactical solution is less detailed than the operational one but it has to be computed in very short time and under imperfect information. The problem is of importance in various applications where tactical and operational planning problems are interrelated and information about the operational problem is revealed over time. This is for instance the case in certain capacity planning and demand management systems. We formulate the problem as a two-stage optimal prediction stochastic program whose solution we predict with a supervised machine learning algorithm. The training data set consists of a large number of deterministic (second stage) problems generated by controlled probabilistic sampling. The labels are computed based on solutions to the deterministic problems (solved independently and offline) employing appropriate aggregation and subselection methods to address uncertainty. Results on our motivating application in load planning for rail transportation show that deep learning algorithms produce highly accurate predictions in very short computing time (milliseconds or less). The prediction accuracy is comparable to solutions computed by sample average approximation of the stochastic program.