(See the key words at the end of this document to find suitable linkages.)
My current interests are in the area of Design for Supply Chains, which is an extension of my past Research and Teaching background in allied fields. Current work is aimed at developing Decision Support Systems to aid the product development process right at the early conceptual design stages. The idea is to prune ineffective designs early and to provide direction to the new product development teams working within the vast conceptual space. These cross-functional product development teams are to be provided with capabilities for handling all downstream constraints within the supply chain (i.e., as the product moves from "cradle to grave") so as to enhance value to the customer. Thus the aim is to provide support at a much larger systemic level by considering all downstream problems within the value chain.
The focus is on enhancing front-end capabilities for helping firms challenge more aggressive supply chain strategies such as Early Design Involvement of Suppliers, Supplier Development, and Inter-organizational cross-functional team participation. The research is aimed at:
This work draws on my earlier research and teaching background in allied fields.
The primary aim of any "Design for X" methodology is to assist the designers (either experienced or inexperienced) by addressing up-front all downstream concerns of a manufacturing process. Computer simulation of the manufacturing process is carried out to aid the design process where necessary.
Most of these design-for-X methodologies, however, rely heavily on compiled heuristics or symbolic knowledge to guide the iterative design process. It is argued that the solution of real world design problems requires design methodologies which draw on all possible sources of design knowledge, whether it be low-level numeric data obtained from a finite element simulation or high-level human-based cognition [1].
Grosse and Sahu [1,2] have developed a system, called Cognitive Symbolic and Numeric Designer (CSN-Designer), for assisting the designer in making intelligent manufacturing and functional based design changes. The CSN-Designer is a computational system, which integrates numerical, as well as compiled and cognitive sources of knowledge, for preliminary design of injection molded components. Some research directions from this background can be pursued as per the following plan:
Plan 1: To extend the above research to other domains where deep knowledge could be obtained from computer simulations of the manufacturing process. It may be possible to extend this concept to other manufacturing domains, such as casting, forging, extrusions, etc. This would enable CSN-Designer to interact with other sources of domain specific knowledge. The main objective, however, would be to strive for domain independence in a semi-automated design environment. This could be achieved by de-coupling the CSN-Designer from the knowledge base by means of a Manufacturing Advisory System [see Plan 2].
Plan 2: Parallel efforts can be undertaken to develop a Manufacturing Advisory System to guide the design while advising on manufacturability, estimating costs and performing automatic process planning. The manufacturing advice could be based on either deep knowledge (as indicated in the above research) or on compiled heuristics accumulated and refined from various individual and organizational experiences. Expert System Technology can be used to capture the relevant manufacturing expertise.
Plan 3: The Manufacturing Advisory System can now be used to de-couple the feature-based CSN-Designer from the domain specific knowledge base thereby resulting in a single computational framework for aiding a wide spectrum of functions from conceptual design to production and maintenance.
Plan 4: The above-cited research also provides a methodology for the integration of continuum-based numerical simulations into a computational system for concurrent design of mechanical components. The technique attempts to develop higher level abstractions from low level numeric data through the help of suitable representations. Any domain specific knowledge embedded in such raw data can be abstracted through similar representations for various front-end (or upstream) applications. The work can also be translated into the business domain to abstract information, from raw data, for effective decision making.
The output from Plans 1 to 4 can help in enhancing the front-end capabilities in any New Product Development initiatives cited earlier. This work incidentally is also influenced by some of the courses that I teach.
My teaching interests are primarily aimed at orienting students along the value chain. I teach them Production and Operations management and then Total Quality Management and finally Supply Chain Management. That is I start by exposing them to issues in the shop-floor and then moving onward to handling organizational issues and then finally to deal with inter-organizational issues within the chain.
The course on Supply Chain Management was developed last year with the help of Dr. Bidhu B. Mohanty of Norfolk State University, USA. I am thankful to him for providing us the latest material on this subject. Dr. Mohanty and I started this work in early 1999 with an intention of developing meaningful interactions with the surrounding industry. We want the industry to understand and adopt the desirable concepts of SCM. We realized that developing a full course would be one of the means to achieve that end. While our attempts to impart this knowledge to the Indian industry is still on, we would like all those who participate in our courses to provide us their valuable feedback and join with us as important change agents.
While imparting the latest literature [3] available in this course to my students, I am also interested in "characterizing" some of the domestic supply chains with the help of student projects. These projects will be helping us identify supply chain constraints, which can be used for enriching the library of features so much relevant for the front-end.
I expect my students to present as much of the work online as possible. Thus projects are mostly administered online (using GroupWare such as Lotus Notes) to enhance gain sharing among work groups.
Dr. Sahu has received funds recently from the AICTE to establish the Computer Aided Product and Process Development Lab for supporting the above research work. He is also looking for collaborative support from other researchers in this field to take some of the research plans forward.
NOTE: IF YOU ARE A RESEARCH SCHOLAR LOOKING FOR RESEARCH OPPORTUNITIES PLEASE CONTACT kaushik_sahu@yahoo.co.in [Click here to see what all opportunities are available in this field.]
1. Grosse, I.R.; Sahu, K. (1994) "Preliminary Design of Injection Molded Parts Based on Manufacturing and Functional Simulations", Advances in Feature Based Manufacturing, (J.J.Shah, M.Mantyla and D.S.Nau, Editors), Elsevier Science B.V., pp:289-313.
2. Sahu, K,; Grosse,I.R. (1994) "Concurrent Iterative Design and the Integration of Finite Element Analysis Results", Engineering with Computers, 10: pp. 245-257.
3. M. Eric Johnson and David F. Pyke "A Framework for Teaching Supply Chain Management" Production and Operations Management, Vol. 9, Number 1, Spring 2000.
Conceptual Design, Product Development, Concurrent Engineering, Quality Function Deployment, Design for Manufacturability, Design for logistics, Design for environment, CAD/CAM, CIM.