Bridging the gap

Citation

(2010), "Bridging the gap", Assembly Automation, Vol. 30 No. 1. https://doi.org/10.1108/aa.2010.03330aaa.002

Bridging the gap

Article Type: Viewpoint From: Assembly Automation, Volume 30, Issue 1

We need change. This is one of the most repetitive phrases I have encountered the past few years. It may be true, as most roadmaps indicate that the research and development (R&D) efforts of the past decade have not resulted in concrete industrial exploitations. Concurrent engineering, flexible systems, reconfigurable ones and simulations to model the processes have not yet given any world region a competitive edge. The problem may be due to the fact that assembly, and in particular the assembly of complex products, has preferably been outsourced. The battle against costs is a difficult one and, if we add the short-term thinking of decision-makers, then one may actually ask if technological advances really do stand a chance. Whatever the reason, it is strikingly evident that the R&D community is fairly stuck in its old trenches as the European Communities (both industrial and academic) all call for sustainability, adaptability and, as Professor Wesskämpfer puts it, a “New Taylorism”. So the big question becomes “how DO we accomplish this?” Could the answer possibly be right there in front of us?

One of the most difficult things for a human being to do is accept change. If we add the possibility that this necessary change may conflict with ones belief in what is required to make progress, then things may get very arduous indeed. Are we desperately clinging to old approaches to R&D and system development? Or are we so committed to solving the issues at hand that we fail to see the “big picture”? There is an old Swedish saying that helps us underline this somehow: “One often doesn't see the forest for all the trees are in the way”. Are we stuck in old ways of doing things? Are we failing to see the holistic picture and obstinately focus on a partial issue? Possibly, because assembly systems are not islands of technology but actually a set of systems that need to collaborate (design, sub-assemble, assemble, test, package, distribute, etc.).

This little fact alone prompted a group of researchers to critically analyse their approach and, instead, take a look at natural systems that proved to be incredibly versatile and adaptable. To be quite honest, the study did not start with a well-grounded intention or structured plan. It all started with a member seeing a photograph in National Geographic (Figure 1) …

Figure 1 Army ants forming a bridge

Ants, one may quite correctly ask, why ants? Let us take a quick look at some simple but solid facts: ants are not intelligent as we humans perceive it. They are not multi-functional, or “flexible” as humans or other higher animals are. Ants are dedicated to specific tasks, simple in nature, quite expendable and extremely goal oriented. They are often hated, hunted, easily terminated and yet they are growing in numbers. In fact, a recent study shows that ants are probably the only other species than humans that actually has achieved a global society: there is a species that has spread itself, through a single colony, to global levels (http://news.bbc.co.uk/earth/hi/earth_news/newsid_8127000/8127519.stm).

Let us return to the picture. It shows ants forming a bridge with their bodies, in order to allow their fellow workers to cross from one branch to the other. Being rather small-brained, this behaviour triggered the group's imagination. Two things were immediately apparent: this action has to do with group (swarms) and goal-oriented behaviours. So the perception of how “intelligence” was being treated changed quite immediately: adaptability may have a lot more to do with dedicated units acting together towards a common, and simple, goal than far-fetched complex algorithms embedded in advanced machinery. And the units are not flexible! On the contrary, they (ants, bees, etc.) are quite dedicated. It is the group behaviour that enables the community to adapt and flourish, not the individual.

A few of you may now be wondering what I am rambling on about. The fact is that, for at least two decades, the R&D community in assembly has focused primarily on the development of multi-purpose, complex, automatic assembly solutions: flexible, reconfigurable machines. Or advanced control systems, or off-line programming systems, etc. So have we been barking up the wrong tree? Does our truly sustainable and adaptable solution actually reside within many dedicated modules that collaborate to achieve changing objectives? Probably, but the context must be made an integral part of the study: tasks and environmental conditions must become active ingredients!

As this was quite controversial to begin with (and still is, right?), the group in question delved into the hidden problem in this equation: the ants' goals do not seem to change often, but production objectives do. So, how does one adapt the group behaviour to rapidly changing goals? How does our group of entities evolve to meet new challenges? The questions are valid because one suspects that, should a major holocaust take place, ants (and not humans) would probably survive and adapt …

Well, let us be honest: the analogies to nature have a limit. At present, however, intelligence does seem to have a direct relationship to both the tasks that are to be solved, and the environment that conditions these tasks. It is therefore plausible to believe that future adaptable, modular systems have the ability to incorporate increasing levels of machine intelligence, depending on the task AND environmental conditions (volumes, TTM, etc.). Which, from a sustainability point of view, makes sense because sustainability requires more than a mechanical or electronic solution: it encompasses social, economic, technological and other aspects. It is multi-disciplinary. It leads us to complex systems.

Consequently, one may propose that future assembly solutions will require a multi-faceted approach in which both internal (tasks) and external conditions (environment) be exploited to assist in the continuous adaptation of the production systems (evolution?). It will basically be a tough new world in which those who can account for these obviously complex systems, in a structured manner, will succeed. And, since we have to move to new roaming grounds continuously, it has to offer us a structured map of the forest and how to get through it without hitting the trees. As ants do, with or without roadmaps.

Mauro OnoriProfessor at KTH, Stockholm, Sweden