The purpose of this article is to discuss how to get “buy-in” from senior or executive management for implementation of Process Analytical Technology (PAT). We must first define PAT:
PAT is a process for building quality into products through a thorough understanding of the manufacturing process, especially Key Quality Attributes (product specific) and Critical Process Parameters (process specific) [1]. PAT generally involves in-line, on-line or at-line measurements at timely intervals to enable the manufacturer to define a “Design Space” [2]. The Design Space is the multidimensional plot of critical process parameters and their acceptable ranges, which define a product that will always be of high quality when the manufacturing process is kept within this Design Space.
Where does the buy-in from senior management come in? It comes in because the instrumentation, personnel, consultants, etc. that are required to implement PAT are expensive. They are generally expensive enough to be designated as “capital projects” by most organizations. The engineers and scientists who are often responsible for implementing PAT generally do not have the authority to sign for the amounts of money that are required to initiate or maintain a PAT initiative. Thus, it becomes a necessity to build a business case for senior management in order to have an effective PAT program. Scientists and engineers are too often caught up in the elegant science of a specific PAT application without doing a careful economic analysis of how the application might (or might not) yield a positive economic return.
Building a business case requires us to be able to talk about Value Added Indicators [3]. Value Added Indicators (VAIs) are tools to evaluate whether an investment is going to give acceptable returns. In today’s business climate, it is not sufficient to simply have a net positive return. The returns must meet or exceed the company’s growth strategy in order to make a convincing business case. The three VAIs we would like to discuss for purposes of this article are:
I. Net Present Value (NPV)
II. Internal Rate of Return (IRR)
III. Return on Investment (ROI)
If these indicators give returns that meet or exceed the company’s growth strategy, it will be easy to build a convincing business case for the implementation of PAT. After discussing the definition of these VAIs, we would like to apply them to a hypothetical business case for illustrative purposes.
Net Present Value (NPV)
Net Present Value is the difference between the Present Value of cash inflows and cash outflows. NPV is used to consider future returns, and discount them to a Present Value. This Present Value takes into account that the returns are generated over a defined period time (typically three to five years) and that inflation will erode some of the value. The future cash flows are discounted into one lump sum, known as the Present Value Amount. The amount applied to obtain the lump sum (the Discount Rate) is set by senior management and the Finance Department. For example, if we anticipate that an investment in PAT will bring in $1M over a five-year period, we might assign a Net Present Value of $800K to that cash flow, using a 20% discount.
The formula for calculating NPV is:Internal Rate of Return (IRR)
IRR is also known as the Economic Rate of Return. IRR is calculated as the Discount Rate that makes the Present Value of all cash flows from a particular project equal to zero. IRR is a measure of the rate of growth that a given project is expected to generate. Obviously, the higher the predicted IRR, the more desirable a specific project would be. Thus IRR is a useful tool for prioritizing projects if several projects are being considered. It should be noted here that IRR is a company specific VAI. This is in contrast to NPV, which is considered to be a universal VAI.
Note that the “$1st yr. investment” in the equation above would be a negative number, as it is an expense. The equation can then be solved for IRR.
Return on Investment (ROI)
Return on Investment is calculated as the NPV divided by the initial investment cost. This is multiplied by 100 to convert to percent:
ROI is analogous to IRR in that the higher the value the more desirable the project. It differs from IRR in that it is a universal VAI, not a company-specific VAI.
The three most important criteria when using the VAIs are:
1. NPV is positive.
2. IRR is greater than the discount rate of the company.
3. ROI should be high enough to make the project worthwhile.
A project could have a positive NPV, yet give an IRR that is less than the discount rate of the company. This would generally not be considered an acceptable return, and thus the project would not be undertaken.
Case Study
Let us now examine how to build a business case for the implementation of PAT. The example we use will be that of a mid-sized pharmaceutical company, and the technology to be employed will be Near Infrared (NIR). This example is only to illustrate the process. Each technology, application and company is different.
The NIR is proposed to be used in three phases as follows:
I. Raw Materials
II. Manufacturing
III. Quality Control (QC)
Our discussion for our business case will focus on only #I, (Raw Materials) above. However, it is instructive to look at how the three parts can be further subdivided with respect to application of PAT:
I. Raw Materials
A. Identification (ID) of Incoming Raw Materials
B. Determination of Physical Properties (Moisture Content, Particle Size)
II. Manufacturing
A. Drying & Milling
B. Granulation & Blending
III. QC
A. Content Uniformity of Drug Product
B. Statistical Analysis of NIR Images of Drug Product
We want to determine the Value Added Indicators for the Raw Materials application of NIR as noted above. This requires some estimation for our Input Variables. First we need to determine how many raw materials are tested in a given time period and what the cost is for that testing.
We assume that the company is testing 400 raw materials per year. We further assume that the average cost of this testing is $95/hr. Testing costs are generally variable. For example, we might have development testing, validation testing, and release testing. The $95/hour value would then be an average of these testing types, assuming that the cost varies for each type of testing.
Next, we will determine our instrument requirements and the associated costs. This can generally be done with a high degree of accuracy, as we would obtain a detailed quote from the instrument vendor for each instrument. We will use two instruments for our business case, a Method Development Sampling System (MDSS) and a Fiber Optic Module (FOM). The MDSS will be used to build identification libraries for the raw materials. This would be done in a Development group. The FOM will be used to perform the identification on raw materials whose libraries have been constructed. This would be done either at the receiving dock or in the raw materials QC group. We will further assume that we can build 20% of the libraries (0.2 x 400 raw materials/year = 80 libraries/year) each year. We can now compile a detailed cost analysis for the first year. Table 1 (below) details the first year cost of NIR implementation for raw materials testing.
| Component | Cost ($US) |
| Method Development NIR Instrument |
77,000 |
| Fiber Optic Module NIR Instrument | 66,000 |
| Commissioning Cost | 9,000 |
| Training | 6,000 |
| Creating Libraries | 15,000 |
| Comparability Protocol | 6,000 |
| Validation of Libraries | 6,000 |
| Vendor Qualification | 3,00 |
| IQ/OQ/PQ | 25,000 |
| TOTAL 1st Year Costs | 213,000 |
Table 1: Total First Year Costs for NIR Implementation
There are some additional input variables that must be added to complete our calculation of the VAIs. We must estimate the cost of testing if we do not implement the NIR testing and continue to use traditional testing in the raw materials laboratories. We must then use our discount rate to convert this cost to a Present Value. Similarly, we must estimate the Present Value of the operational costs for the NIR and the cash flow that the NIR implementation will generate. The estimates for these costs are shown in Table 2 (below).
|
Component |
Cost ($US) |
|
Total expense of traditional testing |
451,000 |
|
PV of traditional testing costs (5 yrs) |
319,000 |
|
PV of Operational cost for NIR (5 yrs) |
49,000 |
|
PV of Cash Flows (5 yrs) |
485,000 |
|
1st year Costs for NIR Implementation |
213,000 |
Table 2: Additional Input Variables for Calculation of VAIs
These values are used to calculate (using an Excel format) the VAIs as follows:
NPV – $272,000
ROI – 128%
IRR – 36%
IRR is a complex formula and is calculated by the Excel program. We used a discount rate of 20% for our calculations. We have already noted in our discussion of VAIs that the ROI is a more generally applicable indicator than the IRR, as the discount rate varies between companies. The ROI is also easier to calculate, although IRR can be done readily using computer software. Note that our NPV is highly positive and our IRR exceeds our discount rate by 16%. The ROI is also quite high. All the VAIs for this hypothetical case study would indicate that this project should be funded.
The chart below shows the Present Value of the cash flows for the five year period, 2007 to 2011. These calculations involve only the use of the NIR instruments for raw materials testing (Phase I). Note that the cash flow increases year to year as a larger number of libraries for the raw materials are completed and used for Identification testing.
The NPV, ROI and IRR would be predicted to increase from 2012 through 2016, as Phase II and Phase III were implemented. However, these phases would require the same type of exercise that we have done here, as new instruments and accessories would be required. For example, a sapphire window might have to be installed on several blenders in order for the NIR to be able to acquire online spectra of a blend.
IV. Conclusion
Due to the capital-intensive requirements of Process Analytical Technology applications, it is generally going to be necessary for a scientist, engineer or manager to be able to make a business case for a specific application. This is done by calculating Value Added Indicators to show the project’s financial benefit to the company. The value of some of the inputs, such as instrument costs, can be calculated very accurately. Other values, such as testing costs, have to be estimated using available data.
References
1. Guidance for Industry, PAT – A Framework for Innovative Pharmaceutical Development, Manufacturing and Quality Assurance; FDA (CDER) September 2004
2. Guidance for Industry, Q8, Pharmaceutical Development
3. Information on NPV, IRR and ROI can be found in any comprehensive business or investing textbook, or on the Internet.
About the Authors
C. Freeman Stanfield, Bir Gujral, and Doug Rufino are with DSM Pharmaceuticals, Greenville, N.C. If you wish to contact the authors, please address any correspondence to: [email protected]
