Search

The societal demand for inspiring and engaging science, technology, engineering, and mathematics (STEM) students and preparing our workforce for the emerging creative economy has necessitated developing students' self-efficacy and understanding of engineering design processes from as early as elementary school levels. Hands-on engineering design activities have shown the potential to promote middle school students' self-efficacy and understanding of engineering design processes. However, traditional classrooms often lack hands-on engineering design experiences, leaving students unprepared to solve real-world design problems. In this study, we introduce the framework of a toy design workshop and investigate the influence of the workshop activities on students' understanding of and self-efficacy beliefs in engineering design. Using a mixed method approach, we conducted quantitative analyses to show changes in students' engineering design self-efficacy and qualitative analyses to identify students' understanding of the engineering design processes. Findings show that among the 24 participants, there is a significant increase in students' self-efficacy beliefs after attending the workshop. We also identified major themes such as design goals and prototyping in students' understanding of engineering design processes. This research provides insights into the key elements of middle school students' engineering design learning and the benefits of engaging middle school students in hands-on toy design workshops.

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ejor.2006.01.010 Byline: S. Kameshwaran (a), Y. Narahari (a), Charles H. Rosa (b), Devadatta M. Kulkarni (b), Jeffrey D. Tew (b) Keywords: Goal programming; e-Procurement; Multiattribute auctions; Multiple criteria optimization; Configurable bids; Business rules; Volume discount cost functions; Bid evaluation problem; Mixed integer linear program Abstract: One of the key challenges of current day electronic procurement systems is to enable procurement decisions transcend beyond a single attribute such as cost. Consequently, multiattribute procurement have emerged as an important research direction. In this paper, we develop a multiattribute e-procurement system for procuring large volume of a single item. Our system is motivated by an industrial procurement scenario for procuring raw material. The procurement scenario demands multiattribute bids, volume discount cost functions, inclusion of business constraints, and consideration of multiple criteria in bid evaluation. We develop a generic framework for an e-procurement system that meets the above requirements. The bid evaluation problem is formulated as a mixed linear integer multiple criteria optimization problem and goal programming is used as the solution technique. We present a case study for which we illustrate the proposed approach and a heuristic is proposed to handle the computational complexity arising out of the cost functions used in the bids. Author Affiliation: (a) Electronic Enterprises Laboratory, Department of Computer Science and Automation, Indian Institute of Science, Bangalore 560 012, India (b) Manufacturing Systems Research Laboratory, General Motors Research and Development, Warren, MI, USA Article History: Received 30 October 2004; Accepted 8 January 2006

In this note, we develop and prove a rule for an upper bound on the optimal number of carriers (with respect to throughput) in a three-workstation closed serial production system with finite buffers, operating under production blocking (blocking-after-service). This system is, in fact, a three server (single server) closed tandem queueing system with finite buffers. Our assumptions regarding service time distributions are nonrestrictive and include the case of iid random variables. Index Terms--Closed queueing networks, discrete-event systems, manufacturing systems, performance comparison, queueing systems.

The hierarchical overlapping condition (HOC) method for solving convex optimization problems has been developed. The HOC uses two or more decompositions that are associated with different partitions of the design variables and constraints. A sufficient condition for the global convergence of the HOC involving the rank of a matrix derived from the Jacobian of the constraints has also been derived. The condition guarantees convergence to a stationary point when the HOC is used in nonlinear optimization problems.