How can Machine Learning Support the Practice of Modeling and Simulation? —A Review and Directions for Future Research

The use of Machine Learning (ML) has achieved a significant momentum across a very wide range of domains. This paper aims to provide a meeting point for discussing the integration of Modeling and Simulation (M&S) with ML. The discussion presents arguments in favour of why and how the M&S practice can avail of ML in different modalities. In this context, the paper reviews key studies published over the past 6 years in main venues including Winter Simulation Conference, SIGSIM PADS, and DS-RT. Further aspects are discussed, which could help reinforce the utilisation of ML in the M&S arena. In general, the study is conceived to foster the presentation of potential ideas and speculative directions towards availing of data-driven knowledge provided by ML.