Human Holon Communications within a High-Mix, Low-Volume Holonic Manufacturing System Organizational communication is fundamental to high-mix manufacturing success. At the human holon level, communications are typically of the form: request, reply and inform. To have dialog amongst collaborative holons, human-system interfaces are required. Interface applications within a cooperative holonic domain are relatively simple, and decision support criteria will look like it comes from a single application.

Cooperative domains are highly distributed, autonomous and cooperative. An example of a strong heterarchy is real-time Internet chat amongst engineers. Heterarchical control is most effectively initiated by a triggering mechanism, e.g., machine breakdown, stockout, etc. for a manufacturing production environment. All users see the same state “real time.” Triggering is initiated after a disturbance or disruption occurs and the response strategy is communication to the entity(ies) capable of mitigating the problem quickly. Decision support would be provided to the holon responding. Decision support is comprised of heuristics to aid in the decision-taking process – a pseudo expert system. The critical importance of mobilizing heterarchy cannot be overstated. To ensure efficiency is maintained, heterarchy increasingly trumps hierarchy as complexity and rate of change increase.

The power of heterarchy can be analogized to the behaviors of the ant kingdom. Simple creatures following simple rules, each one acting on local information. This heterarchical behavior forms the foundation of what is called swarm intelligence, i.e., decentralized self-organizing systems. No centralized control structure dictates how individual holons should behave. The collective behaviors of holons interacting locally cause coherent global benefits to emerge. The triggers and response tactics incorporated act to preserve advice given globally; for example, a production plan.

Consider a situation where a disturbance occurs at a process step. This trigger will cause the local holon experiencing the disruption to inform the production manager holon. The production manager holon will ask for a time estimate to recover from the disruption. If production was suspended, the local holon would have triggered the engineering holon as well as the production manager holon that production was suspended. If the due date cannot be met, the production manager holon will contact customer order service and customer order service may relax the due date or contact the customer directly. The production manager holon may minimally modify the current schedule to minimize the disturbance’s impact on orders in queue yet to be produced.

The rules-of-the-game are embedded in canons which are comprised of heuristics. Canons contain expert advice that facilitates some of the following characteristics for the holons with which they are associated; self organization, system optimization, cooperation, intelligence, forecasting and even considerateness.

Although there is a plethora of information available about automation exhibiting holonic behavior and test cases abound, this is is not a necessary requirement.

R. Michael Mahoney

R. Michael Mahoney is a leading thinker on high-mix strategy, and the author of the college text High-Mix, Low-Volume Manufacturing, published through Hewlett-Packard Press by Prentice-Hall © 1997. He can be found at http://www.Holonyx.com