A two-stage stochastic model for pig production planning in vertically integrated production systems

• Published in: Computers and electronics in agriculture Vol. 176; p. 105615
• Main Authors: Nadal-Roig, Esteve, Plà-Aragonès, Lluís M., Pagès-Bernaus, Adela, Albornoz, Víctor M.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.09.2020; Elsevier BV
• Subjects: Abattoirs; Case studies; Decision making; Hogs; Pig production planning; Pig supply chain; Pig vertically integrated system; Production planning; Stochastic models; Swine; Tactics; Two-stage stochastic programming model; Uncertainty; Windows (intervals); Two-stage stochastic programming model; Pig vertically integrated system; Pig supply chain; Pig production planning
• ISSN: 0168-1699; 1872-7107
• DOI: 10.1016/j.compag.2020.105615

•First multi farm, multi stage and multi period model having the uncertainty in sales prices.•Modelling the pig production process capabilities.•Coordination of the Pig supply chain. This paper focuses on vertically integrated pig companies based on multi-farm systems. The aim is to address tactical decisions to plan production, increase flexibility, improve coordination and overall pig production under the uncertainty associated with future selling price. Decisions to purchase additional piglets and/or rent farms to adapt system capacity were considered. We propose a two-stage stochastic programming model with selling price as a stochastic parameter under a limited time horizon and a case study to illustrate their use. The model maximizes the net revenue of the system by considering a steady piglet production on sow farms and the corresponding animal flow according to growth stage throughout different farms such as breeding, rearing and fattening farms. All-in-all-out management and marketing time window to sell pigs to the abattoir were modeled on fattening farms. The stochastic solution for the case study provides an optimal first stage production plan regarding the purchase of 1016 piglets/week in addition to the 775 already produced beside the renting of rearing and fattening farms taking into account the different scenarios may happen in the future. The model is capable of identifying inefficiencies or bottlenecks in the system. We discuss the value of the stochastic solution of k€1683 compared to the deterministic solution, and concluding the valuable incorporation of uncertainty.