Harvest logistics in agricultural systems with multiple, independent producers and no on-farm storage

• Introduces a solution approach that overcomes challenges found in the literature.
• Presents a novel use of a technique borrowed from the piecewise linearization.
• Introduces a provably optimal algorithm for truck assignment to known loads.
• Computational results demonstrate not only the effectiveness of our approach.

The best way to organize the logistics of harvesting agricultural crops requires considering not only the fact that agricultural commodities in general are highly perishable, but also the fact that the organizational structure of the agricultural system in question can vary from crop to crop and from region to region within a single crop. This paper develops a model for planning the movement of the crop from farm to processing plant for crops satisfying two conditions: (1) there are multiple, independent producers (farmers), and (2) no significant on-farm storage exists. We will also briefly describe three different but economically significant agricultural systems in the United States: sugarcane in Louisiana, sugar beets in the northern areas of the United States e.g. South Datoka, Minnesota, Colorado, and vegetable harvesting for human consumption, and will argue that these systems fit the two conditions of our model. We will also briefly explain why several other significant agricultural systems do not fit these two conditions and hence require alternative modeling techniques. Finally, we demonstrate that the model is computationally tractable by introducing new datasets based upon the sugarcane industry in Louisiana. This choice was driven, not only by the fact that the datasets can be constructed entirely using publically available information on the sugarcane infrastructure in Louisiana, but by the fact that this particular organizational structure also appears in both the sugar beet and vegetable processing industries.