Le 09/11/2018 par Thierry Garaix :

Title: Proactive Workforce Scheduling

Place: Mines Saint-Etienne

Funding: ANR project - Flexible Intelligent Transportation Systems (www.)

Remuneration: 2 400€/months

Duration: 1 year

Skills expected:

Proficiency in C/C++

Proficiency of optimization methods / operational research

Required profile:

Doctor in computer science and/or in industrial engineering specialty optimization.

Supervisors:

Thierry Garaix (garaix@emse.fr, +33 4 77 42 66 41), Mines St-Etienne, Saint-Etienne, France

Marina Vinot (vinot@isima.fr, +33 4 73 40 50 08), ISIMA, Clermont-Ferrand, France

Context:

Home services are a growing sector. On the one hand, the range of services offered widens dramatically (delivery of meals, cleaning, care, help with daily or administrative tasks, etc.). On the other hand the reservation mode of these services evolves and diversifies with the explosion of offers through Uber-like platforms. Thus, the reactivity of these systems to unforeseen events is essential to guarantee the quality of services rendered and their economic viability. All of these issues are grouped under the term: Workforce Scheduling [1].

Subject:

This research aims to develop proactive / reactive approaches to designing robust / flexible planning for the Workforce Scheduling problem [2,3]. Several types of uncertainties are considered: the availability of the drivers, the availability of the customers and the delays in the progress of the initial planning. The unexpected absence of a driver will be the first case study. In this case, it will be necessary to reassign their activities either to another driver or to distribute them among several drivers. In this case, the robustness of a schedule is measured by the deterioration in the value of the objective function produced by the reallocation of an activity (for example, additional kilometers or the increase in working time).

The more complex case of customer availability will be addressed later. The additional difficulty comes from the multiplicity of the sources of uncertainties, as: (1) the variation on the transportation times, (2) the modification of patient care schedules and (3) the arrival of new requests.

Exact and / or approximate methods may be addressed in this postdoc. The exact methods considered are based on stochastic programming and robust optimization [4,5]. The approximate solution methods developped in the project will generalize existing heuristics by redefining the classical operators of the literature.

References:

[1] Castillo-Salazar, J.A., Landa-Silva, D., Qu, R., 2016. Workforce scheduling and routing problems: literature survey and computational study. Ann. Oper. Res. 239, 39–67.

[2] Billaut, J-C., Moukrim, A., Sanlaville, E. Flexibility and Robustness in Scheduling. ISTE-Wiley publishing, London, pp.349, 2008.

[3] Lees-Miller, John D., and R. Eddie Wilson. "Proactive empty vehicle redistribution for personal rapid transit and taxis." Transportation Planning and Technology 35.1 (2012): 17-30.

[4] Gounaris, Chrysanthos E., Wolfram Wiesemann, and Christodoulos A. Floudas. "The robust capacitated vehicle routing problem under demand uncertainty." Operations Research 61.3 (2013): 677-693.

[5] Feillet, D., Garaix, T., Lehuédé, F., Péton, O., & Quadri, D. (2014). A new consistent vehicle routing problem for the transportation of people with disabilities. Networks, 63(3), 211-224.