The context of our work
Let’s be clear about one thing first: We can’t do without you. The subject knowledge that you have, your company’s background knowledge, is key to fast development of your new materials, new analytical methods or new (parts of) your processes. We don’t replace that, we build on it to help you grow it.
Also engineers don’t know what they don’t know. Engineers (mechanical, chemical, etc.) may be even educated at University levels, but were educated in their respective fields of knowledge. Not in all the fields of mathematics. Unfortunately, a typical University education does not (yet) include clear strategies to efficient development. Even post-university Six Sigma training is unlikely to cover all the needs.
What typically happens when a team develops a new product or process, is that the product or process is capable of meeting several of the required quality demands, but not yet all of them. That is normal, as the development is still in an early stage. What is needed now, is finding out if there are combinations of process settings that do make it possible to meet all quality demands. Unfortunately, changing process settings in the attempt to improve one quality parameter will usually negatively impact other quality parameters. Also, changing one process setting may affect other process parts. And since we develop new products or processes, thus develop new knowledge, the available subject knowledge is likely to be insufficient to provide the answers.
As a result, it will be unclear if and how the situation can be optimized. In fact, the problem has changed from an engineering problem into a mathematical problem: how to optimize many process settings towards many quality parameters? Regular engineering experimentation will not be efficient.
A structured rapid development approach will allow the optimization of many process settings towards final products that are good in all aspects simultaneously.
Design of Experiments provides the mathematical structure that is required to allow this simultaneous optimization, and is a key aspect to efficiently develop the new knowledge that is needed. The results of it will give a clear picture of how the (many) different process settings will interact with each other and how they will simultaneously influence the different quality parameters. Computer models can now predict optimal combinations of process settings, to ensure that all quality demands are met. When applied correctly, this methodology provides optimal results in the shortest time frame, and with minimal resources. It is the best scientific approach to speed up complex developments.
Our working process
If you would hire us, what would we do? First we will make sure that at least a basic non-disclosure agreement is covered, as we need to be able to build on your knowledge.
We will initiate a few meetings with the team (managers, engineers and operators) involved in the development at hand. We will collect the already available knowledge. We differentiate clearly between proven knowledge (with data we will check), and claimed knowledge (based on experience, expectation or hunches). We are experienced in this field of information gathering, so it takes less time than you might expect. The resulting overview, a team effort, is often perceived very well by the team ("finally we see the clear picture"). It is the first step in our working process.
The team is now invited to participate in setting up potential strategies for next steps in the development. Here our 20+ years of experience with Design of Experiments in different industries will allow us to guide the team to the strategy that is most likely to be successful, in the shortest time frame, at minimal cost. We will make sure the plan is supported by the entire team, as we consider this a key factor to successful deployment.
Up till now, the only resource investment that the team members have to make, is typically a few meetings of a few hours, spread over 2 or 3 weeks.
The development strategy typically consists of one or more series of experiments that are expected to build the required knowledge to successfully complete the development. The key aspect of the plan is that all process settings (parameters, raw materials, etc.) will be optimized simultaneously towards all quality demands. How this possibly can be done, is covered by our mathematical background and experience. We will make sure that the team understands the basics of such approach, so they can feel comfortable with it.
The experimentation phase of the development typically takes about 1 to 4 months, during which one or some consecutive series of experiments are conducted and analyzed. If the team has expertise in Design of Experiments, most of the preparation and analysis can be done by themselves and we will only supervise. On the other hand, if needed, we can even provide a temporary engineer to do all the required support.
Contact us now to find out if your developments can benefit from this approach.
A course on Design of Experiments is available in English (in-company only) or Dutch, and is organised in cooperation with Avans+, Breda, The Netherlands. For more information on when Design of Experiments is useful, click here.