Tech Trends: Cost Control30 Apr, 2005 By: Arnie Williams
Trim Manufacturing Costs During Product Design
Knowledge reuse management technology recently earned Imaginestics (www.imaginestics.com), a company based in Indiana in Purdue Research Park, a National Science Foundation Small Business Innovative Research Program grant worth half a million dollars.
The grant is for Phase II of a project to develop commercial software that will help engineers design with manufacturing costs in mind—knowing up front what kinds of cost ramifications will result from decisions made in early design stages (figure 1).
Figure 1. New software developed by Imaginestics will read a CAD file and then take the shape through a couple of iterations to determine all the processes and materials needed for casting. The analysis includes cost factors, giving the design engineer more information upfront to make smarter, more cost-effective design decisions.
Today's global competition has forced many companies to focus on their core competencies and outsource all other activities. This trend promises to grow, especially in the United States. The unfortunate result is that manufacturers have little, if any, control over the costs of raw materials, which have become global commodities. As companies outsource more to foreign countries to take advantage of cheap labor, a disconnect between OEMs and suppliers can—and often does—wreak havoc on a company's ability to control product-development cycles and overall costs.
A trend in engineering education exacerbates the problem. At most engineering schools, specialization has led to the elimination of coursework focused on skill-based manufacturing from the core curricula. Instead, these courses are now part of technology programs in which students lack fundamental engineering knowledge, such as that acquired in material and environmental properties classes.
Imaginestics has taken on the daunting task of addressing these trends, which the company views as a basic gap in knowledge between design and manufacturing. Through development and commercialization of software such as i-advisor, the company hopes to transfer manufacturing knowledge that currently resides downstream in the supply chain further upstream to designers and engineers during the early stages of product development. Using i-advisor, the engineers and designers will be able to make cost decisions based on the latest technologies.
Cost-Management DesignGiven that most designers and engineers don't calculate the manufacturing costs of their design ideas as they go along—or can't even if they want to—how would this technology help? The idea is to integrate i-advisor with a designer's CAD system of choice. As parts are created, i-advisor calculates costs in real time in the background, keeping the designer appraised of cost variations that result from design decisions (figure 2). When the design is complete, i-advisor evaluates the part, telling the designer which manufacturing processes are possible, candidates for creating the part, cost estimates for different processes and a list of local suppliers capable of producing the parts using the given processes (figure 3). Further, using the Internet-based ToolingNET technology, also developed by Imaginestics, an RFQ (request for quote) can be sent to participating suppliers.
Figure 2. Through the analysis of such factors as volume, surface and process (green sand—horizontally parted), i-advisor software can calculate associated costs.
During the early proof of concept in Phase I of this project, Imaginestics concentrated on the casting industry, with the support of the American Foundry Association, says Nainesh Rathod, president of Imaginestics. "Based on variables such as volume and surface area," says Rathod, "we were able to tie costs directly to (the removed) material as well as volume and removal sequencing of the process, amongst other cost drivers."
Figure 3. The cost-estimation feature of i-advisor can tell users whether the processes selected are manufacturable according to criteria such as thin sections, surface areas and tolerances.
But in the case of castings and forgings, human decisions enter in. "The ultimate goal would be to reduce the number of cores for casting," Rathod says. "To do that, you need to analyze the distribution of materials—thickness and thinness. Other decisions need to be made, such as where to build in risers and how to reduce the number of risers. The more material you supply in these areas, the higher the costs."
Rathod envisions software that behaves the way a Microsoft Word document does, for example when it puts a squiggly line beneath a misspelled word as users type. If users are designing something in plastic, for example, and introduce a certain feature that eliminates or changes a process—say blow molding vs. injection molding—then i-advisor will alert them to what they've done.
Knowledge Management SuiteImaginestics has already developed i-advisor software to support castings and plastics and has a version for composites, stamping and forgings under development. Other applications the company markets include i-migrate, which automates the task of moving and synthezing content among diverse systems, such as CAD, ERP and PDM; i-compare, which locates content within drawing files based on shape, text, size, author and so on; i-config, a product configurator based on rules and a reuse engine that can create or match design configurations and shows components in an existing configuration; and ToolingNET, an online marketplace for OEMs and suppliers.
"What we mean by knowledge reuse is not about mining data," says Rathod. "It's about how to take data and make it usable at the point of application."
Challenges AheadAs you can imagine, the behind-the-scenes rules to accurately gauge cost and manufacturability factors are complex. Some of the challenges being addressed in Phase II of the Imaginestics project include how to get reliable extraction of shapes from proprietary CAD formats. Rathod notes that because not all design features map to manufacturing features, there is a need to develop an ontology of manufacturing features with intelligent mapping to design features.
Imaginestics hasn't forgotten the "education" gap, either. The company sent an exploratory letter to a number of universities explaining its project, and all have indicated an interest in incorporating the software into their engineering curricula.
Arnie Williams, former editor-in-chief of Cadence magazine, is a freelance author specializing in the CAD industry. E-mail Arnie at email@example.com.