FandMmag.com |

Imagine if, by looking at an acorn, an arborist could predict just how much water, sun and other nutrients it needed to grow into a tree. In manufacturing, that tree is the end product; the nutrients the material, machinery and labor; and the acorn the design—the CAD model. As with acorns, for manufactured goods people have long known what "nutrients" are required: the material, machinery and labor. But the exact amount, both for trees and for manufactured products, remains a mystery until after the fact.

In manufacturing, the measure of the amount that matters is, of course, cost, or cost of goods sold (COGS). Traditionally, industry has taken a "rear-view mirror approach to product costing," says Frank Azzolino, president of Concord, Mass.-based aPriori (www.apriori.com).

Azzolino's company is touting a technology that he says hopes to, in essence, reverse the mirror, to delve into cost estimates early, during design. The technology integrates directly with CAD, incorporates information about potential processes and facilities, and can appear as a kind of "cost-ticker" on the bottom-right of the screen. Cost estimates and detailed analysis can be viewed in a Web browser, too. Want to change the geometry, the material or other part feature? The cost in the corner automatically changes. According to the company, as more design decisions are made, the more accurate the cost estimate in the corner becomes.

It's not as simple as an on-screen ticker, Azzolino says. The estimate weighs in on various factors that the user can tweak, from the cost of different materials to the processes themselves, be it three-axis milling to sheet-metal fabricating and welding, studying the different relationship geometric features have with each process.

"We also examine the information about the facility," Azzolino says, "different routings, different machines and processes, and the feeds and speeds that drive them."

For example, if a part is initially designed for sheet metal, the software allows users to estimate what happens if the same part were injection-molded. It identifies the primary cost drivers, be it tooling, material or other dominant factors. These can be rolled up to include piece-part and assembly-process cost.

The core of this technology lies in advanced mapping techniques of complex interrelationships. According to Azzolino, about a decade ago one large equipment maker wanted to get better at predicting exactly how much a product would cost to make.And so through an industry-academic partnership, university lab graduate assistants and professors cranked the numbers, seeing if they could create "mechanistic models that would allow the design engineer to determine production-level costs in some cases years before production," he says.

The modeling can be applied to all the processes that take place before machines start making parts on the floor. Besides design and manufacturing engineering, this also includes process planning, plant and project management—and purchasing.

Manufacturers don't make products in a vacuum. The mechanistic models look up and down the supply chain, and work well for those with in-depth knowledge of their supplier's cost structure. With globalization on manufacturing's doorstep, Azzolino explains, "local suppliers need to rely on very transparent relationships. They need to do this to survive. It's too easy to go to China, and it may not always be the right solution."

Such transparency in the local supplier base, he says, helps spread efficiencies up and down the supply chain. If an OEM implements cost-cutting measures, they expect the same from suppliers, as many shop owners know all too well.