The prospects are exciting for the drug industry, as considerable benefits and savings can be achieved by working with JMP and its design of experiments feature.
Six Sigma Specialist at NNE A/S
A famous article in The Wall Street Journal expressed it this way: “Pharmaceutical manufacturing techniques lag far behind those of potato chip and laundry-soap makers.” The availability, quality and costs of drugs are today severely limited by the pharmaceutical industry's lack of statistical and scientific approaches in the development and optimization of production methods. The need for streamlined production processes has become more important than ever.
With more than 80 years of experience, Danish engineering company NNE Pharmaplan specializes in filling precisely that need for its clients. NNE Pharmaplan is a leading supplier of systems, consulting and engineering services to the international pharmaceutical and biotechnological industry. It is also a descendent of Novo Nordisk, a world leader in diabetes care and other pharmaceutical products.
Using JMP statistical software from SAS, NNE Pharmaplan provides consulting services for the optimization of production processes based on design of experiments (DOE) using JMP.
“Today, we use a scientific approach for the design and optimization of production processes,” says Per Vase, a Six Sigma Specialist at NNE Pharmaplan. “With JMP, we can evaluate which experiments to conduct in order to find the best possible design and adjustment of the production equipment.
“Previously, we had a non-scientific approach to planning that led to many dead ends, unnecessary experiments and, on the other hand, experiments that should have been conducted but never were,” adds Vase. “Since many factors have significant influence on pharmaceutical manufacturing processes, it was often impossible to see the broader perspective. Using JMP and the DOE feature, the testing becomes more efficient, and we can more promptly decide on the right experiments.”
NNE Pharmaplan believes the scientific approach to process optimization is the right way to achieve greater production efficiency, more consistent quality and less waste, Vase says. “Our goal is to help clients avoid unnecessary costs such as replacing machinery or implementing considerable quality control measures as a result of excessive product waste. Instead we increase their level of process understanding by implementing DOE, so they are able to get more production capacity out of existing equipment and are able to release products based on inline measurements avoiding lead-time increasing QC tests. Creating this process understanding in the pilot phase helps get production up and running to capacity faster from day one after launch.”
At the start of the process planning phase, NNE Pharmaplan engages in a thorough and systematic brainstorming session with the client about the most important aspects of the production that might affect product quality and the overall determination of the quality level for the finished product. Quality standards in the pharmaceutical industry are high but should not be needlessly strict, Vase explains.
“Once we have a clear view of the inputs and requirements for the outputs, JMP comes into play as an efficient analytical solution for controlling the development and optimization of a production process with many variables – a typical scenario in the drug industry. JMP can map out which experiments are reasonable to conduct in order for us to ensure that the final production is optimum,” says Vase.
DOE is iterative and requires a series of designed experiments to find the optimum settings and deliver the required level of process understanding. The first runs have the main purpose of distinguishing between factors that are really critical from those that are only believed to be critical. After finding the really critical few factors, later experiments have the purpose of finding the relation between these and the key outputs of processes and determine the best operating region for the process. Having found the best settings of the critical inputs, confirmation runs are performed to verify the proposed solution. The commencement of the DOE process needs to be planned and resourced several months before the anticipated launch date.
The US Food and Drug Administration (FDA) appreciates the scientific approach to process planning through design of experiments and has encouraged the drug industry to use software like JMP for planning new products.
According to Vase, “The prospects are exciting for the drug industry, as considerable benefits and savings can be achieved by working with JMP and its design of experiments feature.” Although the FDA has been promoting these methods as tools that should be used during the whole product life cycle – from the early development stages to the product being withdrawn from market – NNE is focusing its efforts on implementing DOE at the later stages (pilot or production).
The idea is to have a shorter payback time on client investment. The methods are new for the pharmaceutical industry, so before adopting the techniques on a larger scale (which would include new products that may be four or five years from market launch) it is necessary to prove the value of DOE on a current or soon-to-be-released product where the commercial returns are immediate and more easily measured.
Furthermore, says Vase, “At NNE Pharmaplan, we use JMP daily with customer applications, and it is extremely easy to use. No programming skills are required – the program can simply be used with a mouse.”
By using JMP in the production planning phase, NNE Pharmaplan ensures that its customers gain a better understanding of the scientific approach. Consequently, analysis of clients' existing production lines is frequently also in demand at NNE, and JMP software plays a key role here, as well.
“Companies accumulate many interesting data about their production, but far from all data are used to improve the production,” Vase notes. “JMP collects production data from the different databases, and we use that data to analyze how our customers can improve their production processes. Often, the goal is to avoid excess product waste followed by extensive and expensive quality control and sorting.
“It is quite easy to use JMP for these tasks since it can communicate with all common databases,” says Vase.
Back to Top