The Greening of U.S. Manufacturing

Whether through regulatory force or an effort to deliver on consumers' desires for greener products, an area of increasing importance to manufacturers is the health of the environment with which they interact. A very real fear among many manufacturers is that the resurgence of American industrial activity could be quickly undermined by what some see as unnecessarily harsh environmental regulations. Industry spokespersons are quick to point out that waste and emission treatment is said to cost U.S. industry as much as $75 billion annually. However, a number of efforts are now underway with the aim of maintaining high productivity within manufacturing while fully complying with Environmental Protection Agency (EPA) directives.

Within the state of California, for instance, an ambitious environmental project entails a variety of methods to improve the manufacturing process. This project - the Consortium on Green Design and Manufacturing (CGDM) - was formed in 1993 and is centered at the College of Engineering at the University of California, Berkeley. The CGDM is an industry/government/university partnership designed to develop linkages between manufacturing and design and their environmental effects; it also serves to integrate engineering information, management practices and government policy-making.

Research at CGDM includes:

Development of environmentally conscious manufacturing process models that reflect waste streams, energy utilization, scrap production and process rate.
Development of a process planning model for electro-mechanical subassemblies which analyzes the environmental tradeoffs between dissimilar manufacturing processes.
Analysis of issues in product disassembly, reuse, recycling and disposal.
Development of hierarchies of managerial metrics for measurement of environmental performance by organizations.
Analysis of the use of process-based policy incentives as an alternative to command-and-control regulations.

According to Paul Sheng, director of the CGDM, "We're looking now to develop real-world ways to apply the tools we have developed. And we eventually would like to come out with something we could license for use in the manufacturing arena. But right now, we're just getting all the technologies into place."

Real world uses for CGDM's research have already been established at Sun Microsystems (Mountain View, Calif.), a computer manufacturer. CGDM developed a recycling process for the computer chassis Sun Microsystems uses. Thus far, the company has already incorporated certain aspects of the recycling process in its next-generation products. CGDM is now looking at Sun Microsystems' suppliers in regard to manufacturing waste. "We're modeling the environmental aspects of Sun Microsystems' suppliers to get an idea of how their processes influence design," said Sheng.

The CGDM is also taking part in a three-year project with Ford Motor Co. (Dearborn, Mich.) to study environmentally conscious machining. Sheng noted that beyond their work with Sun Microsystems and Ford, CGDM is also looking into the manufacturing processes behind injection molded parts and printed circuit board fabrication and assembly.

In addition to its work directly in the manufacturing arena, the CGDM has developed the Green Manufacturing Shell - a software tool designed to evaluate the environmental impact of different manufacturing processes. The software achieves this through a multi-objective analysis of different output dimensions such as Process Energy, Process Time, Mass of Waste Streams and Quality parameters.

The software consists of three parts: Green Machining Analyzer, Health Hazard Scoring Module and Materials Database. While the three modules have independent functionalities at this point, the CGDM plans for considerable interaction between them in the future with the final goal of providing robust process planning decisions that incorporate environmental factors. The manufacturing processes currently supported by the Green Manufacturing Shell include drilling, end milling, face milling, broaching and grinding. Access to the software is currently restricted to partners and affiliates of the CGDM.

Emissions Monitoring
Just trying to wade through the morass of regulations can be a daunting task, let alone trying to comply with them. One of the more sweeping requirements of the 1990 Clean Air Act, for instance, places strict limits on the gaseous emissions of manufacturing plants, particularly in regard to nitrogen oxide (NOx), which is believed to be a major cause of acid rain. This has led to a need for manufacturing plants to be able to monitor and predict such gaseous emissions. Although monitoring of this type can seem daunting, neural networks - a computer-based technology that uses adaptive techniques that mimic the way the human brain works - have proven to be particularly well-suited to the task.

Pegasus Technologies (Painesville, Ohio), for instance, has developed a neural network-based system for on-line heat rate improvement and reduction of NOx issuing from its coal-fired furnaces. By optimizing the processes involved in the area of combustion, the nonlinear relationships between a large number of parameters can be correlated. The system has reportedly found better nominal operating conditions and counter-intuitive methods for improving the process even further.

At Eastman Chemical Co. (Kingsport, Tenn.), a neural network was developed from a database of monitored emissions picked up at one-minute intervals over a two-week period. The neural software then trained itself to predict emissions. The end result was a system, dubbed the Software CEM (continuous emissions monitor), which models processes that produce air emissions and predicts emissions based solely on the behavior of those processes. The system meets relative accuracy test audit (RATA) requirements.

A third company, S3 Technologies (Columbia, Md.), has developed a time-delay neural network (TDNN) for NOx and carbon monoxide (CO) prediction in fossil fuel plants. The TDNN model was trained on data obtained from a coal-burning fossil fuel plant. Predictions by the TDNN after training have proven to be significantly more accurate than those obtained by conventional means.

What About Congress?
Despite a flurry of activity on Capitol Hill and many harsh words for the increasingly stringent federal regulations placed on manufacturers over the past 25 years, it's a safe bet to assume that a good portion of existing regulations will stay in place. Numerous polls, even those conducted by pro-business publications, have shown again and again that consumers not only want but are willing to pay more for green products. And it follows that these same consumers who comprise the voting populace are willing to cast their ballots in favor of the politician who helps assuage their environmental conscience.

So, while the more fringe requirements trotted out by the new Congress as examples of overregulation may actually be curtailed somewhat, the manufacturing community should take advantage of the research and pro-environmental applications being put to use in a variety of industries as a means of planning and readiness to comply with current and future environmental regulations.

Click here to return to Table of Contents for the Intelligent Manufacturing August issue.