Sustainable consumption

I am writing this from a technical meeting in Europe I've been attending on manufacturing where the flames of green manufacturing have been flamed and are burning brightly! A separate session on energy efficiency and resource effectiveness saw a group of presentations ranging from more detailed analysis of energy use patterns in production processes (think machining or heat treatment) to more esoteric issues of process planning with energy utilization in mind.

The process planning discussion was interesting. If you are familiar with process planning you already know the complexity of just trying to make sure all machines are used to the fullest extent. Process planning is, basically, how to order the production steps of a product through a number of machines.  It includes how this is optimized to handle the production of a number of different parts (that is, several different sets of parts moving a number of production stations in a sequence - each set of different parts with a different quantity (called batch size)).

Think of the cartoons of production processes shown before here  - a series of boxes linked by transfer mechanisms to move a workpiece from process (box) to process in a sequence. Now think of how a batch of parts of the same component move through this. The first part starts in the first  box where an operation takes place for a set time. Then the part moves to the second box for a second operation and another similar part starts in the first box. With each "cycle" the parts move from box to box until the first part in the batch exits the final box and it is called a "finished product."  Over time, all the parts in the batch move through the production line and the line "falls silent" as the last part moves through the system.

The "falling silent" part is the issue here.

When the next batch of parts (of a different component requiring different times at each of the boxes due to the operations that are needed) starts the production line, the planner has to allow enough time between batches so that the second batch does not "run into" the batch that precedes it. This occurs when the cycle time of some of the boxes is shorter for the second product than for the first one. That is, for a given process applied to a given part, it may require different times to complete the work on a part based on the requirements of the part. And the requirements will change from batch to batch for the parts in the production line.

Further, in such a production line there is always one process that takes longer than the others (called the "bottleneck"). Then, the time in the other steps following the completion of the tasks in that box while waiting for the bottleneck to complete its work is referred to as idle time. The bottleneck may occur at a different station for each batch of parts.

Still with me?

Now, recall the discussion we had in a previous posting on "green at the process level". This identified machines that used energy pretty much independently of the process that was being performed (referred to as "tare heavy") as opposed to machines that used little energy except when performing productive work ("process heavy"). If the production line described above has a lot of stations waiting for a part to appear in order to operate on the part, and the process in the station is "tare heavy", then a poorly planned production process chain will waste a lot of energy while not doing anything productive. Not a desirable situation.

It turns out that a lot of manufacturing processes fall into this category unfortunately for a variety of reasons we won't go into yet.

So, back to the meeting, if one can include in the process planning the consideration of not only delay times (or idle times) in the sequence of starting batches of products (with varying cycle time requirements) but the energy value of that wasted time (do to the machine energy use even if not processing - which will vary from process/machine to process/machine (or box to box in this example), then one could try to find a sequence of production of several batches of products that would insure both minimum production time (or makespan - the time difference between start and finish of a sequence of jobs) and minimum energy used.

This is an industrial engineer's dream problem (and a nightmare to solve).

But, for an existing production facility, for which the processes are well characterized from the energy perspective, this is a realistic goal. A presentation at the meeting by Professor John Sutherland of Purdue University went into some of the details. We can discuss this more at a later time.

So, what about the consumption title of this posting?

At the meeting, following this (and several other) interesting presentations, a discussion started about how if we could just get people to buy more sustainable products, we could produce less overall, and manufacturing would be reduced (although the value of manufactured products would likely be the same or greater) and this would be a better solution than trying to squeeze wasted energy  (or other resources) out of the manufacturing process.

Or as Professor Gisela Lanza of Karlsruhe Institute of Technology put it to me - we need to encourage people not to buy products they don't need with money they don't have to impress people they don't like!

The assembled engineers quickly agreed that we are not into "social engineering" and that this "behavior change" is better left to experts (rock stars, politicians, marketing consultants, other bloggers, etc.)

But, trying to improve the longevity of products by design and manufacturing is something we can aspire to. And maybe the people will follow.

I am encouraged by the fact that Americans seem to be looking for help to do this. Unfortunately they are not getting much assistance from the market place. A recent article posted by Enviromedia commenting on the Federal Trade Commission (FTC) closing its public comment period for its Green Guides states that research that shows 65 percent of Americans would prefer just one seal for green products over the hundreds that are now causing confusion. They note that it is increasingly hard to determine if a product is "truly green" or not based on available information. They are presently overwhelmed with the 350 product certifications that currently exist.

So, the consumer may come around.

In the meantime, there is much to be done to reduce the impact of manufacturing  on the individual process level (and to reduce tare consumption).  This relies on such planning schemes as discussed above. If you have sufficient time between products coming into each box you may actually be able to shut off the process/machine (or essentially put to sleep major components) when the processing is done for that part. Then, if you can restart and warm up the process/machine before the next product appears at that station (box),  to some extent you can "decouple" (a word engineers like to use to mean separate the effect of one thing on the other) the energy optimization problem from the wasted time problem.

And, of course, we can always try to reduce the tare consumption by design of the machine and its control and operation.

We are going to talk more about design and energy efficiency and longevity in the next posting - also motivated by discussions at this meeting.