There has been a lot of talk about going lean and combining it with Six Sigma. Some even doubt that these two can actually work together, and is there proof that they really do?
Personally, I see no reason why the two can’t work together. Both are about streamlining, thereby reducing waste/defect to improve processes for the ultimate customer satisfaction. This scenario is most visible in manufacturing companies. One industry that is working toward lean manufacturing and going Six Sigma is the pharmaceutical indusrty.
Just recently, Pharmaceutical Processing’s Editor-In-Chief Mike Auerbach moderated a live webcast titled Lean Manufacturing, Gaining Efficiencies and Maintaining Compliance on the Plant Floor. The panel included Bill Fitch, VP of Life Sciences, Business & Decision, who discussed how pharmaceutical manufacturers can implement lean and maintain regulatory compliance; Dr. Pankaj Mohan, Manager of Global Process Engineering at Eli Lily & Company, who talked about his firsthand experience in implementing Lean Six Sigma in a pharmaceutical production environment; and Dennis Constantinou, Senior Director of Life Science Industry Strategy at Oracle Corporation, who discussed how information technology can help in a lean enterprise transformation.
I picked up pieces of information from the excerpt that link lean manufacturing with Six Sigma.
Mike Auerbach: Bill, can you give us an example of where process improvement potentially conflicts with compliance?
Bill Fitch: That’s a good question. When we look at the value of making a change to a manufacturing process, we need to evaluate if it is a minor change or major change. If it is a minor change, simply identifying how that change impacts the quality aspects and safety aspects of the medicine, and then providing a report to FDA at the end of the year will be satisfactory. However, if it is a moderate or major change, that would require a supplement to be submitted to FDA. So at the end of the day, you need to evaluate the change that you’re making and the potential challenge that it causes from a commercialization standpoint, especially if it is a product that is already currently on the market.
Mike Auerbach: Dr. Mohan, do you think that value stream mapping is the foundation in all lean environments, and how does value stream mapping link with the DMAIC process?
Dr. Pankaj Mohan: Well, the value stream really sets or fills the hopper that feeds the DMAIC process. So instead of just doing this in a non-structured way, a value stream map can thus provide a very structured and a critical way of thinking through the whole process before you actually begin the process.
Dennis Constantinou: I’d like to comment on that. Before anyone begins any lean initiative, we really need to have an understanding of our business processes. And those business processes need to be highly defined, and not only take it from a business process, but expand that to an entire value plan. We need to be able to do that and have that highly documented and be aware of those processes to the nth degree in order for us to go in and implement any type of solution. Thereby, we can go in and even perform that cost benefit analysis that you were talking about, and being able to do that. Without that, I find it will be very difficult to approach a lean initiative in any kind of structured manner.
Mike Auerbach: Is software validation at the enterprise level and manufacturing data gathering level a stumbling block to implementing IT solutions with lean and six sigma?
Bill Fitch: I would say absolutely not. Does it add additional complexity? Well, sure. Whenever you make a change, you have to deal with evaluating, taking your worst approach and looking at what the current system is, how the system supports the business to make a determination if revalidation is necessary or not. But I think that is something that is inherent in the business and those changes are something that have to be dealt with. Simply being pragmatic about it is the way that you need to approach it.
Dr. Pankaj Mohan: I personally feel that there is no lack of data collection. We can collect collectively in this industry a lot of data. The challenge is to convert this data into knowledge. That is where the various levels of data technology interaction happens. I think that the key challenge is basically taking the data forward and converting it back into useful knowledge that can help lean and can help the process.
Mike Auerbach: How do you reconcile the massive amounts of paperwork generated during validation with the concepts of lean manufacturing?
Bill Fitch: That is a difficult one to reconcile because obviously the very exercise of validation is rigorous development of documentation. So I don’t really know if you can apply the principles of lean to validation. Again, it is a set of discreet activities and different types of documents that have to be put together to show evidence/proof that a system has been developed, tested and functioning and operates for its intended purpose. Unfortunately, you can’t do that with less documents. However, you can take a risk-based approach and be pragmatic about it. In the old days, people would validate all of the functionality of a system. Now the approach is slightly different in that we identify where the regulations impact the functionality or how the system is used. And we’re able to scale validation accordingly, so we don’t do necessarily testing of everything. We don’t conduct the same types of tests on all of the different functionality aspects of the system. So, I guess in a way, I don’t know if I would call that lean, but it is pragmatic.
Dr. Pankaj Mohan: I mean, I personally feel that the validation is crying to be leaned. Many companies have taken a lot of initiatives. I think the key here is to kind of set up the framework with risk-based approach, and then have an efficient documentation trail and execution protocol.
Here’s the entire log of the webcast.
