Beyond the Foggy Supply Chain

Pharma faces myriad challenges in information technology. But, in fact, many of them require the same action: To control costs, a company needs to refine its processes across the board, replacing old spreadsheet-driven processes with sophisticated analytics. That means collecting and analyzing data from a daunting variety of internal and external sources.

To meet these challenges, companies need IT systems that support the ubiquitous flow of data across the organization. But they also need to create and implement new business processes—and in some cases, develop new relationships with suppliers, distributors, and customers who hold key data.

Dennis Constantinou

This business evolution is playing out in various parts of the industry, from R&D to sales management to marketing. But one of the clearest examples comes in sales and operations planning (S&OP), where companies are attempting to replace their old make-to-stock model with a demand-driven approach.

New Approaches

The impetus for change is clear. Approximately five years ago, inventory management agreements (IMAs) emerged. IMAs were devised by manufacturers as a way to align their business objectives with those of the distributors by shifting costs to the most efficient point in the supply chain. But pharma companies aren't good at matching production to demand, so IMAs have had a perverse effect: According to a 2005 report by Pembroke Consulting, pharmaceutical manufacturers added nearly $4 billion in inventory between Q4 2001 and Q4 2004. At the same time, they spent at least $785 million in additional manufacturing operating costs, not counting related expenses such as expedited shipments and the excess capital tied up in inventory.

Arvindh Balakrishnan

In this environment, manufacturers are looking for new approaches to help them unify all business areas—R&D, sales, marketing, manufacturing, distribution, and finance—and truly align supply with demand.

IT plays a crucial role here. With the right systems, companies can:

• Make better forecasts via accurate, easy-to-access historical data and improved collaboration across formerly siloed departments

• Maintain flexibility in production, with the ability to enter new product batches as needed

• Better monitor inventory levels to enable more timely response to fluctuating demand

• Automatically and accurately document all sales and operations processes to ensure regulatory compliance

• Gain complete visibility into end-to-end operations.

Integrated S&OP

Most pharma companies have a fragmented S&OP process. Typically, each department has its own process, storing data on spreadsheets. Departmental plans are not aligned with one another or company objectives. This situation leads to a time-consuming manual process that results in "the forecast," which is typically inaccurate. The supply chain team takes the forecast and figures out how to meet demand—with no thought to the profitability of their decisions. Ultimately, the "approved" plan is filed away, leaving departments to execute based on their own department-specific objectives.

The goal should be to replace this process with one that leads to a single, organization-wide forecast—built on accurate demand signals—which is at the core of an integrated S&OP approach. And the crucial tool in developing this process is integrated IT.

Using integrated IT solutions, manufacturers can perform simultaneous S&OP across multiple distribution and manufacturing facilities, and across various lines of business—including planning, marketing, finance, and manufacturing—within a single planning run, while at the same time accounting for the latest consensus forecast, sales orders, production status, purchase orders, and inventory-policy recommendations. Some systems, for example, allow a direct linkage between sales orders and production batches, enabling users to create a batch reservation for a sales order. When the batch is completed, the reservation for the order line is converted into an inventory allocation and can then be confirmed and shipped. Alternatively, if there are no existing batches planned or in process for the required product, a user can initiate a request to create a batch specifically for that order. Automated workflow notifications apprise order-entry personnel of any changes to the production schedule that may impact their order.

Accurate Demand-Signal Collection

But where does the data come from to enable this sort of forecasting?

Historically, demand-signal-collection models have focused on analysis of distributor orders. But manufacturers have learned through painful experience that these orders provide limited visibility into actual consumer demand. Critically, manufacturers must understand where inventory is in the distribution channel at any given time so they can make accurate, timely adjustments to projections and/or production.

This means that the manufacturer/distributor relationship must be reframed. Progressive manufacturers are pursuing this goal through several means, including pay-for-performance contracts—which place requirements on both manufacturer and distributor.

Some manufacturers are forming bonds with their distributors to capture and share electronic data interchange (EDI) information. Through the efforts of the X12 Committee as well as the Healthcare Distribution Management Association's Electronic Commerce Task Force, several key EDI standards have been developed. These include EDI 852 files (which document product-activity data, including product sold, quantity on hand, and orders) and EDI 867 files (which document product transfer and resale data) to improve demand-signal collection. This gives the manufacturer a better, though not perfect, picture of what the distributor actually has on hand at any given time. It also helps reduce both manufacturer and distributor inventories without having a negative impact on customer order-fill rates.

Other manufacturers are going to the pharmacy level to collect data on the number of prescriptions filled for their drugs. With a significant number of drugs expected to go off patent in the next few years, many pharmaceutical companies are awakening to the reality that they may have to play (at least to some extent) in the generic space, where pharmacy store-level inventory information is crucial.

Collecting prescription-fill data can be complicated; prescription information comes in multiple formats, and not all retailers make it available to manufacturers. However, suitable approximation models are being devised that improve the utility of this information.

In addition, another possible source of demand-signal data could be gleaned from electronic drug pedigrees, which California plans to mandate by 2009. A drug pedigree is a certified record that contains information about each distribution and transfer of ownership of a prescription drug. It records the sale of an item by a pharmaceutical manufacturer, any acquisitions and sales by wholesalers or repackagers, and final sale to a pharmacy or other entity administering or dispensing the drug.

As pharmaceutical manufacturers transform their relationships with distributors, it becomes increasingly essential to have access to IT applications that facilitate trade and partner management and enable advanced analytics. Manufacturers need to effectively capture, analyze, and make decisions based on demand-signal data that is growing in volume and complexity. Spreadsheet-based management cannot fulfill these objectives.

Collaborative Execution

The demand-driven supply chain does not end with the creation of a more accurate forecast. After securing and analyzing information from the distribution channel and using it to develop a more accurate forecast, manufacturers must go back to their distribution partners to determine collective success—a key factor in performance-based contracts. IT systems that enable sophisticated business intelligence and analysis are essential here. For example, manufacturers entering into pay-for-performance contracts with their distributors must be able to set key performance indicators for both parties and to track and analyze performance to determine outcomes and areas for improvement.

The transition from a make-to-stock environment to a demand-driven model requires organization-wide support and a commitment to following the demand signal and a single forecast. It also depends on a solid IT infrastructure that enables the flow of critical operational data across the organization and provides the analytic capabilities needed to make informed and actionable decisions based on that information. These ingredients, when combined and managed carefully, yield a powerful formula for continued success.

John Danese, product strategy director of life sciences at Oracle, contributed to this story.

Dennis Constantinou is senior industry director, life sciences at Oracle. He can be reached at dennis.constantinou@oracle.com

Arvindh Balakrishnan is senior director of life sciences industry business unit at Oracle. He can be reached at balakrishnan.arvindh@oracle.com