Use Simulation, Discipline to Avoid Painful Project De-Scoping

Genuinely critical goals can be identified by one question: "Would we do the project if it did not...?"

 

T

he pharmaceutical and biotechnology industries continue to experience unprecedented pressure to balance the competing interests of manufacturing products "right" and reducing costs. By "right" I mean in accordance with cGMPs, cleanly and with minimal waste, with minimal labor, with ease of product changeover, preferably without hazardous solvents. The list goes on; the point is that an increasing number of non-economic constraints are being applied to manufacturing operations, with a parallel increase in economic pressure.

These pressures can be in direct conflict with the traditional project scoping and early design phases. This inconsistency leads to the painful de-scoping exercise later in the project, and ultimately may lead to dissatisfaction with the overall project.

Typically, a project is initiated by an architectural programming and process conceptual design phase; both can be either formal or informal. Through this process, all of the project needs are identified and approaches to meeting those needs are included in the scope. It is important to note that "need" is a relative term and at this point in a project, "needs" tend to be interpreted very liberally. So, the scope becomes all encompassing and the project often becomes financially unworkable.

Of the number of approaches for avoiding this common situation, some are simple and based on disciplined project management, while others are more sophisticated and intended to optimize the highly technical components of a project.

The first step is to scope the project correctly from the start, which involves identifying the critical goals and constraints. The genuinely critical goals can be identified by the question: "Would we do the project if it did not ,?" If the answer is "yes," then that goal is not critical.

The basic constraints usually involve the owner's budget and schedule issues. A project should begin with these items alone and jealously guard against further additions to the scope. This is perhaps an over-simplification of a real-world scenario, but the degree to which projects often deviate from this common-sense approach is surprising.

Another deviation from good project planning is a failure to plan for cGMP compliance and validation from the beginning. Many projects are scoped and funds are approved, only to learn in later phases that QA will not approve of some aspect of the scope. Invariably, making changes to incorporate the required features leads to either a cost over-run or de-scoping elsewhere. While some issues can be identified only by QA when the project is well underway, many fundamental issues can be identified and included in the scope.

On a more sophisticated level, in meeting competing needs of the manufacturing process itself, computer-based simulation is now being routinely used to model the process for optimization. These simulation tools allow process (or manufacturing) equipment, utilities, material handling and storage space all to be dynamically modeled, balanced and optimized.

Consider the balancing of high-speed filling lines, automatic (or manual) loading and unloading of freeze dryers, freeze drying cycles of various lengths, capping and packaging operations, CIP/SIP cycles, the potential need for cold storage, and many other steps. This illustrates the important nature of balancing equipment and facility attributes--and this is only for one piece of the operation.

The ability to address various "what if" scenarios is another significant benefit of simulation. In the dynamic environment that characterizes our industry, it is the rare project that proceeds from the conceptual phase through design and construction into operation without at least significant questioning of the scope. More common is a steady stream of valid questions: Can the facility manufacture Product X also? What if production requirements doubled? What if Product Z were outsourced?

With dynamic simulation, all of these variables can be analyzed and the questions can be answered quickly. In this manner, the full impact of the change on other products, utilities and facility space requirements can be determined with a high degree of certainty.

Some of these simulation tools provide sophisticated pictorial outputs, showing the operation over time, with tanks visibly filling and emptying; lines running (or not); load/unload carts moving from fill line to freeze dryer; freeze dryers operating, loading or cleaning; and so forth. These models all provide graphical outputs showing production timelines, equipment utilization, utility capacity and other critical measures over a selected period of time. Both planned and random shutdowns are built into the model so that accurate production capacities can be determined.

Without these types of tools, the tendency is to use safety factors to ensure that system capacity is not undersized. This over-design may have been tolerable in the past, but not in the face of cost pressures and the severe competition of the current business environment.

Getting the right balance of capacity, flexibility, risk mitigation and high utilization has long been something of a holy grail to engineers designing pharmaceutical manufacturing operations. Whether you are responsible for manufacturing operations or capital projects, you have probably felt the pain of having to de-scope a project. I am not sure there is a better analogy than "taking candy from a baby," as it may be physically easy to remove scope from a design, but the accompanying pain and suffering may be unbearable.

Carl Bergsten is vice president of pharmaceuticals/biotech for Atlanta-based Lockwood Greene. He can be reached at the company's Somerset, N.J., location at cbergsten@lg.com.

Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments