EVENT

Talk by Prof. Oscar Salviao :"Optimal production plan"

Outline

Date/Time:

January 16 (Monday), 14:00-15:00
17 (Tuesday), 15:00-16:00
19 (Thursday), 14:00-15:00

Place:

National Institute of Informatics
Room 2006 (1/16)
Room 1509 (1/17)
Room 1716 (1/19)

Title:

Talk 1: "Optimal production plan for a manufacturing system with associated recovery process"
Talk 2: "On an Optimal Production-Inventory Plan for a Closed Loop Supply Chain"
Talk 3: "A Problem Based Learning approach based on Bloom’s Cognitive Domain Taxonomy"

Speaker:

Prof. Oscar Salviano (CTI Renato Archer, Campinas)

Abstract:

A time-discrete, constrained, Linear Quadratic Gaussian (LQG) production planning problem is formulatedto develop a production plan with sub-optimal levels of production and remanufacturing for a single product.With the objective to define a strategy of remanufacturing used product, estimated return rates are used to provideproduction scenarios based on this plan.
Nowadays, specific legislation is applied to many industrial sectors regard to the return of used products.Thus, motivated by environmental factors and a shortage of raw materials, partial or total reuse of return productsare a high priority on business's agenda of many companies. In this talk, I will discuss an innovative approach to solvea production planning problem of a dynamic system that includes a reverse channel, with a remanufacturing facility.It is assumed that fluctuations of demand for serviceable products are approximated by stationary normal random variables.Thus, the constrained LQG problem here considered is converted to an equivalent but deterministicthat can be solved by any quadratic programming method from literature, providing, as a result, sub-optimal inventory andproduction plans. Managers can use such plans as scenarios to evaluate future opportunities as costs’ reduction, for instance.
A simple use case will illustrate the approach.

Short bio:

Oscar Salviano Silva is currently a senior researcher at Center for Information Technology RenatoArcher (CTI) and professor of logistics and supply chain management at thePontifical Catholic University (PUC) in Campinas, São Paulo, Brazil. He holds MSc andDSc degrees in the field of automation system at the State University ofCampinas (UNICAMP) in 1982 and 1989, respectively. He has several years ofindustrial experience, coordinated projects involving public and privateinterests; and has been consulting with manufacturing and services companies inBrazil. Currently his main research interests are addressed to three specific aswell as distinct topics: high performance of reverse logistics policies;valuation and pricing of intellectual properties in innovation process; ande-learning technology with base on Bloom Taxonomy associated to Problem BasedLearning.