By Ken Congdon
Over the years, the pharmaceutical supply chain has become the primary target for cost containment measures. While R&D continues to grow and sales overhead remains flat, pharmaceutical companies have repeatedly decreased inventory, closed plants, and increased facility utilization as a way to reduce expenses and boost profits. When it comes to the supply chain, the focus has been on cost containment, efficiency, and simplicity rather than agility and flexibility. Outsourcing has become the norm, which makes much of the supply chain invisible to those responsible for managing it.
One can’t blame the pharmaceutical industry for heading down this path. After all, we took our lead from the automotive, electronics, and apparel industries that have long championed and succeeded using this cost-centric, one-size-fits-all approach to supply chain management. However, if a supply chain interruption halts production of a pair of sneakers or smartphone, consumers are temporarily inconvenienced and company sales may take a brief hit. In pharma, on the other hand, a supply chain-induced drug shortage can literally be the difference between life and death for a patient, and a drugmaker’s reputation can be forever tarnished.
These supply chain risks aren’t news to the pharmaceutical industry. However, many would argue that the system is designed to address these interruptions if and when they arise. For example, most pharmaceutical companies employ a bonus structure based on 95 percent delivery — meaning key personnel receive a full bonus if the drugs they produce are delivered 95 percent of the time. What about the other five percent? Well, the industry feels comfortable that if five percent of a specific drug weren’t delivered, other patent and generic drug manufacturers would be able to fill the gap left by that shortage. This is largely true with many mature, small molecule pharmaceuticals where there are multiple assets of similar chemical makeup in the same class. However, supply chain interruptions become a much bigger threat when dealing with specialty drugs (i.e. medications that are the sole asset and/or class). If a supply chain problem results in production issues with a specialty drug, no one else is equipped to pick up the slack.
This may seem like a small concern today, given that biologics make up less than five percent of total pharmaceutical sales currently. However, according to Andy Skibo, EVP of Operations at MedImmune and Head of Global Biologics and Global Engineering at MedImmune/AstraZeneca, a perfect storm is brewing.
“While bio products account for only 4.7 percent of total sales in the top 100 drugs today, they are poised to account for 52 percent of sales by the end of the decade,” he says. “Currently, BLAs (Biologics License Applications) make up more than 50 percent of the pharmaceutical pipeline, and that figure is expected to grow to 75 to 80 percent by 2020. The pharmaceutical industry doesn’t remotely possess the infrastructure to address this demand. This pending explosion in bio requires supply chain flexibility. Unfortunately, it’s coming at a time where we have the lowest point of capacity, agility, and resilience across the supply chain the pharmaceutical industry has ever experienced.”
Building A Segmented Pharmaceutical Supply Chain
So, what’s the solution to this dilemma? According to Skibo, it starts by abandoning our universal approach to supply chain management.
“Pharmaceutical manufacturers regularly perform quality risk assessments,” he says. “Why don’t we also do supply chain risk assessments? We need to identify supply chain related product risks and be able to quantify those risks in dollars. Based on these findings, we need to develop a segmented model where supply chains are customized based on the needs of specific products.”
In this segmented approach, the supply chains for mature drugs would continue to be managed the way they are today. However, the supply chains for specialty medications would be infused with the flexibility and agility necessary to ensure this sole asset is delivered to every patient on time.
According to Skibo, it’s important for the pharmaceutical industry begin this practice of supply chain risk segmentation immediately. There’s good reason for this urgency. Even though the explosion in biologics may still be a few years out, creating the capacity necessary to handle this demand can take several years (e.g. it can take five years or more to build a new manufacturing facility). With this in mind, MedImmune/AstraZeneca is already testing this segmented model of supply chain management internally, and Lance Minor, the company’s VP of Network Strategy and Business Operations, is the man heading up this effort.
“Supply chain segmentation starts by gaining an understanding of the variability of the process and demand of a product and comparing that with cycle time and the flexibility of your network,” says Minor. “From that calculation, you can determine those products that require some type of flexibility through inventory and network design.”
According to Minor, you typically see a need for network redundancies with new product launches and with products that have longer manufacturing cycle times. Longer cycle times generally result in slower response rates and are riskier due to the high number of process steps involved. In short, more process steps create more opportunities for failure.
AstraZeneca initially performed a supply chain risk assessment on a single product — Synagis, a prescription medication used to prevent a serious lung disease caused by RSV (respiratory syncytial virus) in high-risk children. The company performed an in-depth multivariate Monte Carlo simulation of the typical five key product parameters in the commercial and operational space, such as manufacturing process success, dose, yield, demand variability, and clinical probability of success. The results of these simulations provided AstraZeneca with a fairly broad portfolio perspective of overall demand for Synagis and gave the company an early warning to potential weaknesses in their supply chain for the product.
This exercise revealed that several supply chain adjustments — including yield improvements, cycle time reductions, and inventory builds — were necessary to shore up production of this essential, life-saving medication. “One of the first things we realized was we needed to increase capacity and white space in our bulk facility,” says Minor. “We accomplished this by reducing cycle times in changeovers and run rates, which essentially doubled our output at this location.”
Furthermore, the simulations showed AstraZeneca that it needed to add a linkage within its network that would allow the company to go from one bulk site to an alternate drug product site if necessary — providing added supply flexibility. This move has already proven valuable as capacity constraints at AstraZeneca’s CMO are requiring the company to get this alternate supply chain validated sooner than originally anticipated.
Finally, the company also looked for areas where it had wait times in raw materials and supplies and made inventory changes to ensure it could maintain the stability and staging of new runs should a change in schedule occur. “Through our simulations, we realized that simply bumping up our inventory by three months could essentially lower our risk of shortage to zero,” says Minor.
“These collective capacity expansion efforts have been invaluable,” says Minor. “Without these shifts, we could have lost a couple of years of product supply potentially worth billions, but more importantly, ensuring supply of sole-asset in class medicine.”
Communicate Changes To Internal/External Stakeholders
Of course, alterations to something as crucial to pharmaceutical operations as the supply chain aren’t made overnight. Managing this change requires the input and cooperation of several key stakeholders throughout the enterprise. According to Minor, the process starts by convincing your CFO that specific supply chain modifications are necessary.
“You need to identify the higher inventory target levels required to meet demand flexibility and outline the necessary costs in network changes to ensure supply,” says Minor.
These conversations should be had with other functional teams throughout the organization. For example, you need to ensure the procurement department has visibility into demand variability, so their metrics are aligned with the need for flexibility. Similarly, you need to work with the forecasting team to ensure this variability is reflected in both short-term and long-term product forecasts. In addition, since demand forecasts can often be driven by dosing swings, it’s important to work with the clinical team to understand the dose ranges they are testing on a product and combine upside demand with upside dose.
Supply chain change management doesn’t end with your internal teams, but extends to external suppliers as well. These partners are an important part of your network, but are often where supply chain visibility is obscured. It’s vital that you work with these suppliers to avoid production interruptions.
“We educate our suppliers on our own demand variability so they can best plan their supply chains to accommodate our needs,” says Minor. “Furthermore, we accept the fact that supplier initiated changes are often inevitable. However, we don’t want to learn about these changes at the point they are being made or after the fact. We work with our supplier to ensure we’re involved in the conversation before a change is made so we can influence the final decision and proactively adjust our operations in light of this change.”
Further Application & Automation Of Supply Chain Modeling Needed
While most of AstraZeneca’s supply chain risk assessment and segmentation activity has been focused on Synagis, the company’s goal is to apply this methodology to its entire product line. “Our plants are multi-product facilities,” says Minor. “Therefore, if you have variability in one demand, it can affect the ability to supply in another.”
Minor also hopes to add a level of automation to the company’s supply chain risk assessment and segmentation process in the near future. Currently, AstraZeneca relies on Excel-based Monte Carlo simulations to gain an understanding of demand variability. However, once these calculations are made, running these findings through “what if” scenarios and suggesting potential solutions is a largely manual process. Minor hopes to eventually integrate these models with the company’s SAP software and other forecasting programs so that the system would actually self-diagnose and automatically present a set of suggested solutions.
“In my mind, achieving this type of automation isn’t that complicated from a conceptual standpoint,” says Minor. “There may be some complexity from an IT perspective to actually link the systems. However, we have cars that can drive down the road by themselves today. We ought to be able to have an algorithm that helps us understand what potential supply chain solutions are. Even if this automated system suggests a better solution just one percent of the time, it would be worth the investment.”