Scheduling models in farm management: a new approach.

Three operational planning models to calculate schedules for an arable farm are examined. These models are a linear programming model, a dynamic programming model and a simulation model. They are examined at different levels of aggregation and relaxation in a retrospective way. Also a probabilistic dynamic programming model is examined. A new algorithm is developed for scheduling farm operations. This algorithm which is heuristic (and comes from the field of Artificial Intelligence), is far more efficient and only a little less effective than the three mentioned models. The new model, called FLOS (= Farm Labour and Operations Scheduling), is probabilistic and makes use of the weather forecast and soil moisture and grain moisture models to calculate future workability. With this new model, several tests are executed including a practical test on an arable farm of the RIJP (Lake IJssel polders Development Authority). Around the heuristic FLOS, a Decision Support System called SCHEMA (= SCHEduling Multiple Activities) is developed