Environmentally Conscious Design and Manufacturing with Input Output Analysis and Markovian Decision MakingEPA Grant Number: R825345
Title: Environmentally Conscious Design and Manufacturing with Input Output Analysis and Markovian Decision Making
Investigators: Olson, Walter , Pandit, Sandar , Sutherland, John
Institution: Michigan Technological University
Current Institution: Michigan Technological University , University of Toledo
EPA Project Officer: Karn, Barbara
Project Period: October 1, 1996 through September 30, 1999 (Extended to December 30, 2000)
Project Amount: $180,000
RFA: Technology for a Sustainable Environment (1996) RFA Text | Recipients Lists
Research Category: Sustainability , Pollution Prevention/Sustainable Development
The objective of this program is to create a methodology that constructs a state space model of a product or process system which, when used as a Markovian chain, predicts the long term usage and flow of materials as well as the transient effects of changes made to the system. Leontif input/output model technology is used to capture the state of the product/process. This model permits an analysis of static changes which helps identify targets which have the most effect on the waste streams and the product/process environment. Alternatives are evaluated to provide the parameters to be used in a Markovian decision making process to predict the dynamic industrial system changes. These, in turn, allow a decision maker to optimize a decision of which improvement projects to select, in what order to execute them and to predict the effects that multiple improvement projects will have.
This program is a three year effort which will be applied in a large (>2500 employees), a medium (<2500 employees but >250 employees), and a small (<250 employees) manufacturing company. The initial effort is to collect data and build the input/output model for each company. Software will be created to support this effort. This software will be ported to the World Wide Web. The model will then be augmented with analysis and decision techniques to achieve the goals outlined above.
The project is jointly funded by NSF and EPA. NSF funding will support the model development and validation effort. EPA funding will support identification of candidate industrial facilities, data collection, and development and testing of model software.