R&D Is Like a Jigsaw Puzzle-You Have To Put the Pieces TogetherByRaymond D. Matkowsky
A few years ago a European passenger train suffered a terrible accident that resulted in the derailment of a number of cars and the loss of several lives. The initial part of the catastrophe was caused by a broken wheel. The wheel fracture in of itself would not have caused a serious accident. However, at the time of the break, the train was travelling through an underpass. The broken wheel caused the car to sway excessively and hit the abutment holding up the roadway above. This caused the derailment of the car and every car behind it. The combination of these factors resulted in this catastrophic loss.
A multi-factor effect within an industrial example was a product that would give off an objectionable odor during storage. It turned out that the fumes of two essential ingredients of the product combined at certain temperatures to form the objectionable odor. There are products on the market now that prevent something like this from happening. At that time there wasn’t. In this particular instance, the answer was to substitute for one of the components.
These situations are not unlike what researchers, industrialists, and academics face in projects all the time. However, it is said that 80% of projects fail. In my opinion, they fail because much of the time there is more than one factor contributing to an outcome and most people are just looking for that single factor. If we look deeper we may increase the above productivity dramatically. You have to remember though, finding the right answer starts with asking the right questions. Start your questions by determining the complexity of your problem.
Most problems will fall into one of three categories or a combination of the three:
Single Factor
Many researchers and engineers look for that single factor. In my experience, only the most simplistic of problems can be solved by modifying a single factor. Consequently, most projects fail to achieve what was hoped for.
Independent Multiple Factors
Sometimes, you have two or three factors that act independently. Each one of these factors affects the product to some degree. In this situation, I have a preference for a research technique called “An Analysis of Variance (AOV).” An Analysis of Variances is a statistical technique to compare the differences in means across different groupings. It will tell you the relative strength of each factor and their positive or negative direction. An AOV can be time consuming. However, in a successful AOV, you will spent more time upfront but less time overall.
With the above information, you now can concentrate on the factors that most contribute to the outcome. You can learn how to balance these factors in order to get the outcome you want.
Synergistic Effects
Occasionally you will run into factors that do not affect your product individually or do not affect it as strongly by itself as happens in the presence of another component. Again an AOV will help. An AOV will sense any synergistic affects present, tell you the direction of those affects, and the strength.
Overlooked Factors
It is very common to overlook some factors. Your staff is a factor that should not be overlooked. Over the years, they probably built up a routine as to how they do things. It is very likely that there will be no affect. This, however, should be checked out.
In the same token, your equipment can be a factor. How fast does your steam jacketed kettle heat up? Are there any hot spots? Are you mixing your ingredients well enough? Do you reduce your moisture content sufficiently and fast enough? There are many more possible questions.
Emergence of Clear Picture
Jigsaw puzzles usually come in a box with a picture of how the pieces fit together. You have a picture and the shapes of the individual pieces as a guide. With the development of a new product you don’t have a guide. You have an idea of what you want, but you have no guide to get you there. An AOV becomes your guide.
A word of caution! You still need an experienced manager that knows what is important to keep and what is not. R&D is always a compromise of properties.
Do you have any comments or other suggestions, please share them with your fellow readers. Email me at rdm@datastats.com.
|
