Combining harvesting operation optimisations using strategy-based simulation

Modelling and simulation assist in decision support or planning activities by allowing efficient exploration of multiple scenarios in a situation where testing in a real setting is impractical. This exploration is often done by varying numerical parameters in the model such as physical dimensions or speed in order to find the optimal configuration. However, for certain problems, in order to find optimal solutions it is beneficial to vary the algorithms that are used to implement the behaviour of the model. For example, when calculating optimised routes for harvesters and other vehicles in a harvest operation, the choice of optimisation algorithms is an important part of the problem. Traditional modelling and simulation techniques do not allow us to vary algorithms across simulations effectively. In this paper, we address this issue by applying the strategy pattern from software engineering to the construction of a formal model that enables different combinations of harvest optimisation algorithms to be analysed effectively. This approach can be generalised to other planning activities where multiple algorithms need to be considered.