Dynamic predictive scheduling and line balancing in industry 4.0

  • Posted on: 26 February 2019
  • By: BABOLI
  • Updated on: 26 February 2019
Type recrutement: 
Post-doc
Section: 
27 (Informatique)
61 (Génie informatique, automatique et traitement du signal)
Entité et lieu: 
INSA de Lyon, Laboratoire LIRIS
Urgent ?: 
oui
Détails: 

The main objective of this postdoctoral position is to study the problem and innovative solutions for Dynamic predictive scheduling and line balancing in the context of industry 4.0, based on reel time information and their possible application in the Volvo Group’s factory. These solutions must not only take into account the constraint of assembly line, but also other constraints concerning part replenishment of line, kitting, sequencing of parts, transportation and replenishment of parts from external and internal suppliers (End to End Supply Chain). Some of these constraints are the hard constraints as respecting the available space in existing building, mix model assembly line principle, and some others can be modified, as external sequencing or internal kitting. The innovative solution must allow increasing the line First Time Through (FTT) for final assembly line where diesel and electric vehicle are assembled on a same production line.

In the first step of this project, you will start by studying previous research on scheduling of assembly line and understanding their problem. A literature review on line balancing, scheduling and dynamic modification of these planned activity using real data and information come from shop floor. For example: dynamic decision, dynamic organization, dynamic line balancing, dynamic scheduling, etc. must be studied to a better understanding of subject and possible contribution.

In a second step, focus will move to identification of main data and constraints in order to propose a new method or alternative solutions for line balancing and scheduling of a customize mass production. The main challenge for Customer mass production come from the variation of operation time from a product to the next one in the one hand, and several uncertainties (demand, capacities, delay, breakdown, etc.) on the other hand.

The data and information concerning the products and production systems of several factories of Volvo group: the Blainville factory (Final Assembly line), other Volvo Groups components factories (engines, transmissions …), and suppliers must be studied and analyzed. The methods come from data analytics (descriptive, predictive, prescriptive) have to be used.

Finally, mathematical modeling and optimization, simulation and digital twin could be used to study the validity of the proposed solutions.