Optimization as a tool for decision making in small-scale farming systems

Abstract

Objective: to review the mathematical optimization methods used as a decision-making tool in small-scale farming systems.

Design/methodology/approach: the present study consisted in performing a search for scientific articles in SCOPUS and ScienceDirect using the following keywords: optimization, models, dairy systems, small scale.

Results: it was found that linear programming is a method used to minimize or maximize linear functions subject to equality or inequality constraints. Non-linear programming aims to find the optimum of a function of various non-linear variables. The method called simulated annealing is a meta-heuristic search algorithm for global optimization problems. Finally, the genetic algorithm method differs from a classical derivative-based optimization algorithm (linear and non-linear programming) in two main ways, and generates a population of points in each iteration; the best point of the population approaches an optimal solution and selects the next population using a calculation that employs random number generators.

Study limitations/implications: there is scarce scientific literature reporting the development of mathematical models of this type that allow simulating and supporting decision-making to prevent significant increases in greenhouse gas emissions from livestock production.

Findings/conclusions: Worldwide, various mathematical models have been developed to estimate greenhouse gas emissions from livestock production, although optimization models have been used only to simulate strategic management in small-scale dairy systems.