|
March 2000
Engineer developing software
to resolve design arguments
WEST LAFAYETTE, Ind. – Conflicts between engineers and other professionals designing facilities such as factories may one day be anticipated and resolved automatically by computers, in effect eliminating human egos and fallibility from the process.
Facility design work typically involves diverse types of professionals, such as engineers, architects, construction experts and various managers. Sometimes these experts, who may be in far-flung offices around the world, are unknowingly headed toward major conflicts that will slow or stall the project.
"Companies design and redesign their facilities all the time," says Shimon Nof, a professor of industrial engineering at Purdue University. "Often people argue, especially when they are physically remote from each other."
Nof is working to develop a "facility description language" and other computer tools aimed at preventing design conflicts automatically. He will discuss the research during the Conference on Computer Simulation and Artificial Intelligence, Feb. 16-18, at the Universidad Panamericana in Mexico City.
He also will present research findings on the overarching issue of trying to reduce information overload for employees whose daily routines revolve around "e-work," which is any work that is done via a computer network.
"The premise is that without effective e-work, the potential of emerging and promising electronic work activities, such as virtual manufacturing and e-commerce, cannot be fully materialized," Nof says.
In the specific area of design conflicts, typical disagreements emerge over the proposed layout of the facility. Industrial engineers may argue with facility planners about the arrangement of machinery on the production floor. The width or layout of aisles in a factory or warehouse may restrict movement of vehicles and equipment, such as forklifts and robots. Engineers responsible for the design of manufacturing facilities may lack certain details about safety regulations, resulting in future conflicts with safety design experts. Problems also arise between the designers and their customers.
The experimental computer tools, which incorporate "neuro-fuzzy" programs that endow computers with the ability to learn, are showing promise in resolving such conflicts.
"You might say, 'But that's what people do anyway. Somebody has to solve the conflict,'" Nof says. "Unfortunately, people are not very good at conflict resolution. Communication among people, as interesting as it is, is not very productive for work objectives. People get frustrated. We have to prevent the participants from being frustrated and motivate them to continue what they are doing."
The aim is to create software tools that not only identify conflicts in advance but also describe fine details about the nature of those conflicts to better pinpoint a resolution. So far, experimental results have shown that such software tools are capable of eliminating the "dependency on humans to execute critical design tasks" needed to resolve conflicts that arise in facility design, Nof says in a research paper about the work.
Computer support for conflict resolution is not new. Some companies have used it in designing new products. What is new is the idea of using computer-supported conflict resolution for the design and redesign of facilities, Nof says.
The broader issue is using software tools to improve the efficiency of e-workers. Because they are linked via computer networks to each other and to sensor-laden facilities and equipment, they have a wealth of data at their fingertips.
Too much information, however, can lead to dysfunction. "We are at the point where we have too much computer technology and too much information," Nof says. "We get so much information, just to find out what we don't need takes time that could be better spent on something else."
So the Purdue engineer has been developing software "agents" and protocols, which function behind the scenes in parallel with human workers to improve the efficient flow of e-work. The agents and protocols lighten the human workload by performing certain complicated and specialized functions automatically, says Nof, who will discuss that work during the conference in Mexico City.
"We already have agents," Nof says. "When you use a printer you have an agent in your computer that takes care of all the necessary steps to send the file that you want to print to the printer. So, what we have here is a parallel activity, which is very productive. The printer is printing while you can keep writing.
"Protocols and agents are both tools to work in this parallel layer, so that they do things automatically that we don't want to do or that we cannot do."
Sources: Shimon Nof, (765) 494-5427, nof@ecn.purdue.edu
Writer: Emil Venere, (765) 494-4709, evenere@purdue.edu
Purdue News Service: (765) 494-2096; purduenews@purdue.edu
ABSTRACT
Systematic Resolution of Conflict Situations
in Collaborative Facility Design
Marco A. Lara (Doctoral Student) and Shimon Y. Nof
This article introduces a new method to resolve with computer support conflict situations that typically occur in collaborative facility design (i.e., layout design, equipment selection, automation planning, and material flow design). The method encompasses rational execution of pre-ordered conflict resolution approaches (direct negotiation, third-party mediation, incorporation of additional parties, persuasion, and arbitration) successfully used in solving human disputes. It incorporates principles for prevention of conflict perpetuation and escalation to improve design performance, and applies computer-based learning to improve usefulness. A graph model for conflict resolution is used for the first time in computer-supported collaborative design to facilitate conflict modeling and analysis, one of the incorporated principles for prevention of conflict perpetuation and escalation. The performance and usefulness of the new method have been validated by implementing its conflict resolution capabilities in the Facility Description Language (FDL), a computer tool for collaborative facility design, and by applying FDL-CR, the resulting version of FDL, to resolve the four typical aforementioned conflict situations. The results found in this research are: 1) Incorporation of principles for prevention of conflict perpetuation and escalation improved the effectiveness of the method; 2) Implementation of computer-based learning increased the usefulness of the method; and 3) Integration of conflict detection and resolution resulted in an increased effectiveness of the facility design process.
To the Purdue News and Photos Page
|
Purdue News
