Post-doc offer
Forum 'Emplois' - Sujet créé le 2016-07-16
Cher.e.s collègues,
Dear colleagues,
Je vous prie de trouver ci-après une offre de stage post doctoral de 10 mois sur les problèmes d'optimisation combinatoire dans les systèmes de production sous incertitude, à pourvoir à Angers.
Please find below a 10-month Post-doctoral proposal entitled "Combinatorial problems of production systems optimization under uncertainty" in Angers, France.
Bien cordialement,
All the best,
Postdoctoral research position in Operations Research
Combinatorial problems of production systems optimization under uncertainty
Andre Rossi, Alexandre Dolgui, Evgeny Gurevsky
July 9, 2016
This postdoctoral fellowship is a part of the ADLACI project, funded by the «RFI AtlanSTIC 2020» program of the french region «Pays de la Loire», and dedicated to the optimal design of production systems under uncertainty.
The design of production systems is an important industrial challenge that includes several issues dealing with optimization aspects, named in the literature as balancing problems. In general (see, for example, the survey Scholl and Becker, 2006), they consist in a partition of the set of all necessary production tasks among machines (or workstations)
with respect to a given production goal and subject to some technological restrictions such as precedence, cycle time, exclusion, inclusion or space constraints. One of the important goals at the production systems design stage is the anticipation of the task time variability in order to construct a robust system configuration for a long term usage.
Since 2012 (see Gurevsky et al., 2012, 2013; Rossi et al. 2016), we study and popularize an alternative robust approach for evaluating production systems in such a situation. This approach is based on a specific indicator, called as the stability radius. Given a feasible production system configuration with already assigned tasks, it is calculated as the maximal
amplitude of the deviations of the uncertain task times from their nominal values for which the system admissibility remains respected. The most practical advantage of using this indicator comparing with stochastic and fuzzy approaches consists in the fact that there is no need to possess reliable historical data on the variability of processing task times.
Design a production system configuration with the greatest stability radius is a new difficult combinatorial optimization
problem that has been recently introduced and studied in our work Rossi et al. (2016) for simple assembly lines and represents an important object of study from both practical and algorithmic points of view. One of the goals of the ADLACI project enrolls the perspectives of that paper and will be dedicated to the optimal design of assembly lines with extra large task time variations that can even be greater than the cycle time. In response, the only technological solution, but still very expensive, is the installation of parallel workstations with duplicated tasks that conducts us to consider new complex industrial optimization problems aiming to minimize the number of parallel workstations and seeking the trade-off between the corresponding stability radius to be maximized.
The object of this Post-Doc proposal is to develop efficient heuristic and exact methods for the resolution of this new problem as well as to study the computational complexity of some combinatorial optimization sub-problems that can be appeared in. It will be achieved by studying its particular structure in order to investigate, inter alia, good quality
upper/lower bounds on optima, efficient dominance properties and appropriate branching techniques.
The successful candidate will be located in LERIA (Laboratory of Computer Science) of the University of Angers (France) and is expected to have good operational research and multi-objective optimization skills as well as excellent programming background in C/C++. A previous experience with CPLEX or Gurobi, and knowledge of assembly lines balancing or mono-dimensional bin-packing problems would be appreciated as well.
Compensation: Between 1850 and 2100 euros net per month (depending on the candidate experience).
Contract duration: 10 months from October 2016 to July 2017 (may vary by one month).
Please send a CV, a motivation letter and two references to andre.rossi@univ-angers.fr, alexandre.dolgui@mines-nantes.fr, evgeny.gurevsky@univ-nantes.fr
Bibliography
Gurevsky, E., Battaia, O., Dolgui, A., 2012. Balancing of simple assembly lines under variations of task processing times. Annals of Operations Research 201 (1), 265-286.
Gurevsky, E., Battaia, O., Dolgui, A., 2013. Stability measure for a generalized assembly line balancing problem. Discrete Applied Mathematics 161 (3), 377-394.
Rossi, A., Gurevsky, E., Batta¨?a, O., Dolgui, A., 2016. Maximizing the robustness for simple assembly lines with fixed cycle time and limited number of workstations. Discrete Applied Mathematics 208, 123-136.
Scholl, A., Becker, C., 2006. State-of-the-art exact and heuristic solution procedures for simple assembly line balancing. European Journal of Operational Research 168 (3), 666-693.
Dear colleagues,
Je vous prie de trouver ci-après une offre de stage post doctoral de 10 mois sur les problèmes d'optimisation combinatoire dans les systèmes de production sous incertitude, à pourvoir à Angers.
Please find below a 10-month Post-doctoral proposal entitled "Combinatorial problems of production systems optimization under uncertainty" in Angers, France.
Bien cordialement,
All the best,
Postdoctoral research position in Operations Research
Combinatorial problems of production systems optimization under uncertainty
Andre Rossi, Alexandre Dolgui, Evgeny Gurevsky
July 9, 2016
This postdoctoral fellowship is a part of the ADLACI project, funded by the «RFI AtlanSTIC 2020» program of the french region «Pays de la Loire», and dedicated to the optimal design of production systems under uncertainty.
The design of production systems is an important industrial challenge that includes several issues dealing with optimization aspects, named in the literature as balancing problems. In general (see, for example, the survey Scholl and Becker, 2006), they consist in a partition of the set of all necessary production tasks among machines (or workstations)
with respect to a given production goal and subject to some technological restrictions such as precedence, cycle time, exclusion, inclusion or space constraints. One of the important goals at the production systems design stage is the anticipation of the task time variability in order to construct a robust system configuration for a long term usage.
Since 2012 (see Gurevsky et al., 2012, 2013; Rossi et al. 2016), we study and popularize an alternative robust approach for evaluating production systems in such a situation. This approach is based on a specific indicator, called as the stability radius. Given a feasible production system configuration with already assigned tasks, it is calculated as the maximal
amplitude of the deviations of the uncertain task times from their nominal values for which the system admissibility remains respected. The most practical advantage of using this indicator comparing with stochastic and fuzzy approaches consists in the fact that there is no need to possess reliable historical data on the variability of processing task times.
Design a production system configuration with the greatest stability radius is a new difficult combinatorial optimization
problem that has been recently introduced and studied in our work Rossi et al. (2016) for simple assembly lines and represents an important object of study from both practical and algorithmic points of view. One of the goals of the ADLACI project enrolls the perspectives of that paper and will be dedicated to the optimal design of assembly lines with extra large task time variations that can even be greater than the cycle time. In response, the only technological solution, but still very expensive, is the installation of parallel workstations with duplicated tasks that conducts us to consider new complex industrial optimization problems aiming to minimize the number of parallel workstations and seeking the trade-off between the corresponding stability radius to be maximized.
The object of this Post-Doc proposal is to develop efficient heuristic and exact methods for the resolution of this new problem as well as to study the computational complexity of some combinatorial optimization sub-problems that can be appeared in. It will be achieved by studying its particular structure in order to investigate, inter alia, good quality
upper/lower bounds on optima, efficient dominance properties and appropriate branching techniques.
The successful candidate will be located in LERIA (Laboratory of Computer Science) of the University of Angers (France) and is expected to have good operational research and multi-objective optimization skills as well as excellent programming background in C/C++. A previous experience with CPLEX or Gurobi, and knowledge of assembly lines balancing or mono-dimensional bin-packing problems would be appreciated as well.
Compensation: Between 1850 and 2100 euros net per month (depending on the candidate experience).
Contract duration: 10 months from October 2016 to July 2017 (may vary by one month).
Please send a CV, a motivation letter and two references to andre.rossi@univ-angers.fr, alexandre.dolgui@mines-nantes.fr, evgeny.gurevsky@univ-nantes.fr
Bibliography
Gurevsky, E., Battaia, O., Dolgui, A., 2012. Balancing of simple assembly lines under variations of task processing times. Annals of Operations Research 201 (1), 265-286.
Gurevsky, E., Battaia, O., Dolgui, A., 2013. Stability measure for a generalized assembly line balancing problem. Discrete Applied Mathematics 161 (3), 377-394.
Rossi, A., Gurevsky, E., Batta¨?a, O., Dolgui, A., 2016. Maximizing the robustness for simple assembly lines with fixed cycle time and limited number of workstations. Discrete Applied Mathematics 208, 123-136.
Scholl, A., Becker, C., 2006. State-of-the-art exact and heuristic solution procedures for simple assembly line balancing. European Journal of Operational Research 168 (3), 666-693.