A parallel Branch-and-Fix Coordination based matheuristic algorithm for solving large sized multistage stochastic mixed 0–1 problems

On the time-consistent stochastic dominance risk averse measure for tactical supply chain planning under uncertainty

“…In the presence of uncertain parameters, different approaches for solving nonlinear separable mixed 0-1 problems can be found in the literature in the two-stage and multistage settings. A recent review of decomposition algorithms is presented in [3], most of the algorithms are intended for problem solving with moderate model dimensions. For bigger instances, some types of scenario cluster decomposition approaches can be used, such as Branch-and-Fix Coordination [3,6], two-stage Lagrangean decomposition [12], Progressive Hedging algorithm [27] and multistage cluster Lagrangean decomposition [20], among others.…”

“…A recent review of decomposition algorithms is presented in [3], most of the algorithms are intended for problem solving with moderate model dimensions. For bigger instances, some types of scenario cluster decomposition approaches can be used, such as Branch-and-Fix Coordination [3,6], two-stage Lagrangean decomposition [12], Progressive Hedging algorithm [27] and multistage cluster Lagrangean decomposition [20], among others. For instances with very large dimensions, such as real-life STSCP instances, matheuristic approaches should be used, as the algorithms that belong to the stochastic nested decomposition methodology, see [2,14,22,58].…”

“…Observe the high number of cuts that could be appended to the root node r-rooted subproblem (as many as nite 3 ∈Lr |S 1 |). So, a mechanism should be considered for removing non-active cuts.…”

Call for Papers Issue 1/2020

“…Expectations and business opportunities that result from these developments include the potentials of real-time decision making (e.g., for IT security infrastructure protection and financial stock trading), solving computationally hard optimization problems (e.g., in production and logistics) (e.g., Gendron and Crainic 1994;Aldasoro et al 2017), and analyzing huge volumes of data acquired from sensors, mobile phones, and social networks (e.g., for social network analysis, fraud detection and business analytics) (e.g., Wang et al 2016;Zhang et al 2016). The analysis of such data has already started attracting new machine learning approaches.…”

High Performance Business Computing

“…They further allow moving from offline analytics to realtime analytics. Finally, in the fields of management science and operations research, computing clusters have been used to execute parallel algorithms for a broad set of both theoretically and practically relevant problems in scheduling, logistics, and further application fields of optimization and simulation (e.g., Gendron and Crainic 1994;Aldasoro et al 2017).…”