BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
TZID:Europe/Paris
X-WR-TIMEZONE:Europe/Paris
BEGIN:VEVENT
UID:599@cran.univ-lorraine.fr
DTSTART;TZID=Europe/Paris:20241003T130000
DTEND;TZID=Europe/Paris:20241003T140000
DTSTAMP:20241120T113344Z
URL:https://www.cran.univ-lorraine.fr/events/reunion-sdf-phm2-7/
SUMMARY:Réunion SdF-PHM2
DESCRIPTION:"A Parallel-Machine Learning framework to tune metaheuristics f
 or advanced manufacturing scheduling problems"\, Hanser Jiménez (Postdoc\
 , European project MODAPTO\, CRAN/UL)\nAbstract:\nMeta-heuristics (MH) hav
 e become a de facto approach to find approximate solutions for complex sch
 eduling problems. However\, since the quality of solutions provided by the
 se methods is highly sensitive to the value of their parameters\, tuning p
 arameters is a key and challenging step to guarantee a good performance. T
 unning MHs is not trivial since it is in turn dependent on the complexity 
 of the problem at hand and the available time to perform such procedure. I
 n the context of real-world manufacturing processes\, the specific charact
 eristics of such processes give place to complex scheduling cases\, which 
 turn MHs into expensive-to-evaluate functions for candidate settings needi
 ng to be tested. Such characteristics include the high interaction of semi
 -finished goods and operations in their respective bills of material\, as 
 well as the specific constraints of manufacturing operations that need to 
 be balanced.\nIn this talk\, we propose a Bayesian-Optimization (BO)-based
  framework supported by parallel computing techniques to perform MH’s tu
 ning for manufacturing processes. The proposed framework captures the spec
 ificity of manufacturing processes in a training phase by learning the imp
 act of MH’s parameters on the business key performance indicators. By do
 ing so\, the framework can be used to find a near-optimal parameter settin
 g able to produce efficient schedules for new cases once it is trained. Th
 e proposed framework configures as a managerial tool that can be integrate
 d with existing Advanced-Planning-and-Scheduling software that use MHs as 
 their underlying models.
CATEGORIES:Séminaires projet SDF-PHM2
LOCATION:FST - AIPL\, 745 Rue du Jardin Botanique\, Villers-lès-Nancy\, Fr
 ance
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=745 Rue du Jardin Botanique
 \, Villers-lès-Nancy\, France;X-APPLE-RADIUS=100;X-TITLE=FST - AIPL:geo:0
 ,0
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:DAYLIGHT
DTSTART:20240331T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
END:VCALENDAR