"We aren’t integrating to SAP, and we aren’t doing MES, but we are doing an integration project,” said Joe Musiak, associate director of global automation for Biogen Idec, as he began to detail his company’s ambitious, multi-site integration plans. Biogen makes multiple biologic products in plants in Massachusetts, North Carolina and Denmark and, “We needed to be much more flexible,” Musiak said.
He threw a very large stack of paper on the table. “This is the amount of paper we need to make a batch,” he said. Then, holding up two or three sheets of paper, he added, “When this project is done, we’re hoping that this will be the amount of paper we need to make a batch.”
“We actually did the same project three times, at three of our plants.” Biogen Idec’s Joe Musiak explained the company’s effort to unify and integrate automation practices at its various plants.
“All of our sites are totally DeltaV,” he continued. “However, all of them were designed and built at different times with different engineers and different integrators. Our manufacturing processes and SOPs are heavily dependent on paper, but all of our process requirements are pretty much the same except for scale.”
The problem, simply stated, is that the company was not leveraging the full value of automation, Musiak said. Users felt that the four facilities were pulling further apart in terms of process and automation capability. Differing levels of automation were in turn causing batch-record development differences. Manual operations were labor-intensive and contributed to operator errors. Finally, Biogen Idec had switched to competency-based training, Musiak said, but this effort “was not aligned with automation or batch records.”
In fact, he said, of installing a new chromatography system, “We actually did the same project three times at three of our plants. We needed to do something different.”
Musiak explained that his vice president of manufacturing paired him with a co-leader from manufacturing to ensure top-level buy-in from both engineering and manufacturing, and has served as the project cheerleader from the beginning. The team met and set some achievable goals, including a significant improvement in operating efficiency, a reduction in the cost of the DeltaV software development effort and a reduction in the size and complexity of documentation.
“We made sure that what we were developing was a user-driven plan, not an automation-driven plan. There are times when I think something is neat, but if the users don’t think it is, we don’t do it.”
During a three-month planning period, the team took interviewed 30 to 40 users and defined the current state of automation at Biogen Idec. They introduced the concept of a global automation library, developed a “batch record rule book,” and developed a strategy to achieve the project’s business goals in 30 months. The final, overarching goal of the project is to define the optimal future state of automation and have each site independently harmonize to that state by the end of 2010.
The team began with a prototype media prep project and was soon able to demonstrate success, including a 72% reduction in manual data recording, a 100% reduction in manual calculation, a 58% reduction in total operator actions required to run the operation, and a 53% reduction in the total number of pages of documentation required.
The results of the prototype further showed a reduction in batch-record size and complexity, a reduction in code development time, a reduction in time taken to validate code and a reduced burden of re-training. It also revealed the need for a standardized interface, which allows for a transferable workforce.
The project began in March 2008 and is intended to run for thirty months.