Algorithmically-guided postharvest protocols by experimental combinatorial optimization

The application of search and learning to experimental domains, where the objective function cannot be accurately simulated, but rather requires a measurement in real industrial settings, lies in the focus of this study. We consider the problem of devising treatment protocols for fresh cucumbers, whose quality rapidly deteriorates once being harvested, by considering the combinatorial space of possible postharvest practices. The overall target is to prescribe a combination of treatments, with specified activation levels, that minimizes the cucumbers' quality loss after 4 weeks in two storage environments: 10°C and 20°C. This study engaged with a postharvest laboratory with industrial settings to research and develop a sequential experimentation procedure, in a closed feedback-loop fashion, and subject to strict budget and timeline constraints. The laboratory measurements comprise the assay of color, stiffness and mass, as well as external blemishes - in both harvest and post-4-weeks points in time. Their deviations constitute the aggregated objective function that undergoes minimization for both temperatures. After formulating the optimization problem, we outline our approach and report on the attained results. The obtained protocols significantly outperform the best-known human reference practice, and their nature is visualized and analyzed. Finally, we mention the impact and outlook for industry.

[1]  Christian Fikar,et al.  Bi-objective optimization of e-grocery deliveries considering food quality losses , 2021, Comput. Ind. Eng..

[2]  Ofer M. Shir,et al.  Sequential experimentation by evolutionary algorithms , 2021, GECCO Companion.

[3]  V. Rodov,et al.  Effects of Compostable Packaging and Perforation Rates on Cucumber Quality during Extended Shelf Life and Simulated Farm-to-Fork Supply-Chain Conditions , 2021, Foods.

[4]  Leroy Cronin,et al.  A universal system for digitization and automatic execution of the chemical synthesis literature , 2020, Science.

[5]  Jakob Bossek,et al.  Evolving Sampling Strategies for One-Shot Optimization Tasks , 2020, PPSN.

[6]  Nan Shan,et al.  Comprehensive Characterization of Fruit Volatiles and Nutritional Quality of Three Cucumber (Cucumis sativus L.) Genotypes from Different Geographic Groups after Bagging Treatment , 2020, Foods.

[7]  A. Khan,et al.  Pre-storage Application of L-arginine Alleviates Chilling Injury and Maintains Postharvest Quality of Cucumber (Cucumis sativus) , 2019, Journal of Horticultural Science & Technology.

[8]  I. Kahramanoğlu,et al.  Improving Postharvest Storage Quality of Cucumber Fruit by Modified Atmosphere Packaging and Biomaterials , 2019, HortScience.

[9]  Edward S. Spang,et al.  Food Loss and Waste: Measurement, Drivers, and Solutions , 2019, Annual Review of Environment and Resources.

[10]  J. Panigrahi,et al.  Starch glucose coating-induced postharvest shelf-life extension of cucumber. , 2019, Food chemistry.

[11]  Ofer M. Shir,et al.  Predict or screen your expensive assay: DoE vs. surrogates in experimental combinatorial optimization , 2019, GECCO.

[12]  Judea Pearl,et al.  The seven tools of causal inference, with reflections on machine learning , 2019, Commun. ACM.

[13]  A. Flammini,et al.  The Future Challenges of Food and Agriculture: An Integrated Analysis of Trends and Solutions , 2019, Sustainability.

[14]  D. Migut,et al.  Postharvest Ozone Treatment of Cucumber as a Method for Prolonging the Suitability of the Fruit for Processing , 2018, Ozone: Science & Engineering.

[15]  I. Nasef.  Short hot water as safe treatment induces chilling tolerance and antioxidant enzymes, prevents decay and maintains quality of cold-stored cucumbers , 2018 .

[16]  Ofer M. Shir,et al.  Sequential experimentation by evolutionary algorithms , 2018, GECCO.

[17]  M. Geyer,et al.  Postharvest treatments of fresh produce , 2014, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[18]  Xiao-Jiang Feng,et al.  Fundamental Principles of Control Landscapes with Applications to Quantum Mechanics, Chemistry and Evolution , 2014 .

[19]  Bing Chen,et al.  6-Benzylaminopurine alleviates chilling injury of postharvest cucumber fruit through modulating antioxidant system and energy status. , 2013, Journal of the science of food and agriculture.

[20]  R. Anurag,et al.  Effect of modified atmosphere packaging and storage conditions on quality characteristics of cucumber , 2014, Journal of Food Science and Technology.

[21]  Douglas B Kell,et al.  Scientific discovery as a combinatorial optimisation problem: How best to navigate the landscape of possible experiments? , 2012, BioEssays : news and reviews in molecular, cellular and developmental biology.

[22]  Thomas Bäck,et al.  Mixed-integer evolution strategy using multiobjective selection applied to warehouse design optimization , 2010, GECCO '10.

[23]  Weiru Liu,et al.  A survey of formalisms for representing and reasoning with scientific knowledge , 2010, The Knowledge Engineering Review.

[24]  David W. Corne,et al.  Noisy Multiobjective Optimization on a Budget of 250 Evaluations , 2009, EMO.

[25]  Jacob D. Feala,et al.  Search Algorithms as a Framework for the Optimization of Drug Combinations , 2008, PLoS Comput. Biol..

[26]  M. Kasım,et al.  UV-A treatment delays yellowing of cucumber during storage , 2008 .

[27]  Joshua D. Knowles,et al.  ParEGO: a hybrid algorithm with on-line landscape approximation for expensive multiobjective optimization problems , 2006, IEEE Transactions on Evolutionary Computation.

[28]  T. Nilsson.  Effects of ethylene and 1-MCP on ripening and senescence of European seedless cucumbers , 2005 .

[29]  Thomas Bäck,et al.  Metamodel-Assisted Evolution Strategies , 2002, PPSN.

[30]  Andy J. Keane,et al.  Metamodeling Techniques For Evolutionary Optimization of Computationally Expensive Problems: Promises and Limitations , 1999, GECCO.

[31]  Alain Ratle,et al.  Accelerating the Convergence of Evolutionary Algorithms by Fitness Landscape Approximation , 1998, PPSN.

[32]  J. Dennis,et al.  A closer look at drawbacks of minimizing weighted sums of objectives for Pareto set generation in multicriteria optimization problems , 1997 .

[33]  Hans-Paul Schwefel,et al.  Evolution and optimum seeking , 1995, Sixth-generation computer technology series.

[34]  C. Wandrey,et al.  Medium Optimization by Genetic Algorithm for Continuous Production of Formate Dehydrogenase , 1995 .

[35]  C. Wang,et al.  Chilling-Induced Ethylene Production in Cucumbers (Cucumis sativus L.). , 1982, Plant physiology.