Valeria Borodin

Handling uncertainty in agricultural supply chain management: A state of the art

Given the evolution in the agricultural sector and the new challenges it faces, managing agricultural supply chains efficiently has become an attractive topic for researchers and practitioners. Against this background, the integration of uncertain aspects has continuously gained importance for managerial decision making since it can lead to an increase in efficiency, responsiveness, business integration, and ultimately in market competitiveness. In order to capture appropriately the uncertain conjuncture of most agricultural real-life applications, an increasing amount of research effort is especially dedicated to treating uncertainty. In particular, quantitative modeling approaches have found extensive use in agricultural supply chain management. This paper provides an overview of the latest advances and developments in the application of operations research methodologies to handling uncertainty occurring in the agricultural supply chain management problems. It seeks to: (i) offer a representative overview of the predominant research topics, (ii) highlight the most pertinent and widely used frameworks, and (iii) discuss the emergence of new operations research advances in the agricultural sector. The broad spectrum of reviewed contributions is classified and presented with respect to three most relevant discerned features: uncertainty modeling types, programming approaches, and functional application areas. Ultimately, main review findings are pointed out and future research directions which emerge are suggested.

A quality risk management problem: case of annual crop harvest scheduling

This paper presents a stochastic optimisation model for the annual harvest scheduling problem of the farmers’ entire cereal crop production at optimum maturity. Gathering the harvest represents an important stage for both agricultural cooperatives and individual farmers due to its high cost and considerable impact on seed quality and yield. The meteorological conditions represent the deciding factor that affects the harvest scheduling and progress. Using chance-constrained programming, a mixed-integer probabilistically constrained model is proposed, with a view to minimising the risk of crop quality degradation under climate uncertainty with a safe confidence level. The chance-constrained optimisation problem is tackled and solved via an equivalent linear mixed-integer reformulation jointly with scenario-based approaches. Moreover, a new concept of -scenario pertinence is introduced in order to defy efficiently the probabilistically constrained problem complexity and time limitations. From the practical standpoint, this study is aimed at helping an agricultural cooperative in decision-making on crop quality risk management and harvest scheduling over a medium time horizon (10–15 time periods).

Predictive modelling with panel data and multivariate adaptive regression splines: case of farmers crop delivery for a harvest season ahead

This paper investigates a harvest-season level unbalanced panel data (PD) of farmers crop delivery for monitoring the gathering activity and for aiding to support reception and storage decisions making of an agricultural cooperative. To achieve these purposes, the fitting and the prediction of the daily farmers crop delivery quantities were realised based-on the total expected quantity of the whole harvest season, the daily volume of precipitation and the amount of sunshine. In order to capture and extrapolate data patterns, both the PD regression and the multivariate adaptive regression approaches were implemented and tested for a real life agricultural cooperative case study. The obtained results exhibit an accurate predictive modelling of the farmers crop delivery behaviour for harvest seasons ahead.

A discrete event simulation model for harvest operations under stochastic conditions

This study presents an application of stochastic discrete-event simulation modelling for harvesting, transportation and storage activities of one grain and oilseed agricultural cooperative. Gathering the harvest represents an important stage for both agricultural cooperatives and individual farmers, that requires an efficient management in order to ensure a high production quality and yield. For the purpose to take into account the intricacy and the dynamic behaviour of the studied system, the model proposed here, considers various inherent heterogeneous parameters, such as: daily meteorological uncertainty, loss queuing networks, farmers contractual delivery policies, etc. This paper applies discrete and stochastic simulation techniques in order to analyse and evaluate the performance of the cooperative supply chain system. Moreover, it enables to investigate alternative configurations and strategies of its operations for an eventual supply chain redesign.

A Decision Support System for Efficient Crop Production Supply Chain Management

This paper presents a decision support system for efficient and responsive crop production supply chain management, which captures harvesting, transportation and storage activities, through taking a holistic perspective as a whole. Crop production represents one of the most significant stage in agricultural sector for both agricultural cooperatives and individual farmers, due to its high cost and considerable impact on crop quality and yield. Whilst considering the dynamic behaviour and the intricacy of the studied agricultural system, combined discrete event simulation and optimization approaches, scenario analysis and performance measurement tools are implemented for supporting decision making for cooperatives and individual farmers, respectively. Likewise, it enables to examine and evaluate alternative system re-configurations and strategies for an eventual supply chain rethinking or redesign. The decision support system proves to be particularly responsive and effective when applied to a real life agricultural case study.

A multi-step rolled forward chance-constrained model and a proactive dynamic approach for the wheat crop quality control problem

Handling weather uncertainty during the harvest season is an indispensable aspect of seed gathering activities. More precisely, the focus of this study refers to the multi-period wheat quality control problem during the crop harvest season under meteorological uncertainty. In order to alleviate the problem curse of dimensionality and to reflect faithfully exogenous uncertainties revealed progressively over time, we propose a multi-step joint chance-constrained model rolled forward step-by-step. This model is subsequently solved by a proactive dynamic approach, specially conceived for this purpose. Based on real-world derived instances, the obtained computational results exhibit proactive and accurate harvest scheduling solutions for the wheat crop quality control problem.

An Interval Programming Approach for an Operational Transportation Planning Problem

This paper deals with an interval programming approach for an operational transportation problem, arising in a typical agricultural cooperative during the crop harvest time. More specifically, an interval programming model with uncertain coefficients occurred in the right-hand side and the objective function is developed for a single-period multi-trip planning of a heterogeneous fleet of vehicles, while satisfying the stochastic seed storage requests, represented as interval numbers. The proposed single-period interval programming model is conceived and implemented for a real life agricultural cooperative case study.

Analyzing different dispatching policies for probability estimation in time constraint tunnels in semiconductor manufacturing

In semiconductor manufacturing, new technologies impose more and more time constraints in product routes, i.e. a maximum time between two (often non-consecutive) operations. The management of Time Constraint Tunnels (TCTs, combining multiple time constraints) in high-mix facilities is becoming more and more challenging. This paper first recalls an approach for estimating the probability that a lot at the entrance of a TCT will leave the TCT on time. This approach relies on a list scheduling algorithm using a dispatching policy with random components. Three dispatching policies are presented. Computational experiments on industrial data comparing these policies are discussed. Perspectives are drawn to extend the approach and support decision making.

A Decision Support System for managing Line Stops of Time Constraint Tunnels

In a High-Mix/Low-Volume (HM/LV) 300mm wafer fab, several hundred product routes are active, each with dozens of Time Constraint Tunnels (TCT), the management of which has become a real challenge. In order to help addressing the difficulties it poses, a Decision Support System (DSS) dedicated to managing Line Stops of Time Constraint Tunnels (TCT) is presented in this paper. The proposed DSS is able to detect Line Stops of TCTS, as well as exhibiting the relevant information for the management of the whole set of TCTs in a format that is understandable by a human. After a three stage integration in production, the tool industrialization has been successfully completed.