Open AccessEstimating and tuning adaptive action plans for the control of smart interconnected poultry houses
Author(s) -
Darlan Felipe Klotz,
Dalcimar Casanova,
Marcelo Teixeira
Publication year - 2021
Language(s) - English
Resource type - Conference proceedings
DOI - 10.5753/sbiagro.2021.18381
Subject(s) - action (physics) , action plan , computer science , control (management) , plan (archaeology) , variable (mathematics) , production (economics) , artificial intelligence , mathematics , history , ecology , mathematical analysis , physics , macroeconomics , archaeology , quantum mechanics , economics , biology
In poultry farming, the systematic choice, update, and implementation of periodic (t) action plans define the feed conversion rate (FCR[t]), which is an acceptable measure for successful production. Appropriate action plans provide tailored resources for broilers, allowing them to grow within the so-called comfort zone, without waste or lack of resources. Although the implementation of an action plan is automatic, its configuration depends on a specialist, which tends to be inefficient and result in variable FCR[t]. In this project, the specialist's perception is reproduced, to some extent, by computational intelligence. By combining deep learning and genetic algorithm techniques, we show how action plans can adapt their performance over the time, based on previous well succeeded plans. We also implement a network infrastructure to replicate our method over distributed poultry houses, for their smart, interconnected, and adaptive control. A supervision system is provided as interface to users. Experiments using real data suggest an improvement of 5% on the performance of the most productive specialist, staying close to the optimal FCR[t].