I had promised last year to focus some time on describing best practices in the HP Supply Chain and how they helped making the company more responsive. Today I want to focus on one, called Procurement Risk Management (PRM) that has been widely publicized and received a number of awards. My "google-meter" actually found 633 references to "procurement risk management at Hewlett-Packard company", which, you will recognize, is quite a number.

Let me start by focusing your attention on a document titled "Procurement Risk Management at HP", written by the team who developed the methodology in the first place. In 1999-2000, HP was faced with significant price increases and an availability shortfall for Flash memory. To assure future availability of Flash memory, HP decided to enter in a binding long-term contract with a major supplier. The uncertainty of the future price and availability and HP's own demand uncertainty made specifying the terms and conditions very difficult. Following questions had to be addressed:

• What should we pay for flash memory over the next few years and how should we structure our payments
• How much should we buy and how should we structure delivery terms
• How long a horizon should the contract cover and when is the best time to sign the agreement
• What provisions should be included to secure compliance.

The PRM program was launched in August 2000 to develop and standardize methods of addressing these types of questions. It has since become a cornerstone strategy for procurement at HP and is widely used within the company. It benefits HP, its suppliers and customers.

Over the last five years, based on financial-engineering approaches used on Wall Street, HP has developed:

• A framework to quantify the impact of product demand, component pricing and availability uncertainty on revenue, costs and profits
• Software tools to support the risk management process
• A rigorous procurement risk management process to proactively manage procurement uncertainties and risks

The PRM framework measures uncertainty using a scenario approach and involves the measurement of uncertainties associated with buying commodities and the management of these risks using structured contracts. Uncertainties cover demand, price and availability over time, and for each a low, medium and high scenario is developed. In some situations HP uses a complementary approach, called BRAIN (Behaviorally Robust Aggregation of Information in Networks) developed by HPLabs to gain high quality pricing estimates for example. Correlations between demand, price and availability uncertainties are also captured. Once the procurement uncertainties have been modeled, the current strategy can be analyzed to measure the risks involved.

The uncertainties are managed using structured contracts with suppliers. These contracts are binding commitments between HP and the supplier with complex combinations of quantity and pricing terms. Quantity terms include fixed and flexible quantity contracts and percent of total-available-market (TAM) while pricing terms include discount-off of market price, fixed price, price-caps and price-floors.

Once having identified how much will be sold with "certainty" (high probability), the "certain" demand in components is covered by fixed quantity contracts, while flexible contracts are used to satisfy likely but "less certain" demand. Unlikely demand may still be addressed through spot purchases.

Where used, this approach has helped HP reduce supply chain costs between 15 and 20% , and resulted in a cost saving of 445M$ in the period 2001-2006. Incremental benefits include:

• Materials cost savings, lowering the supplier demand risk while enabling him to cut costs through more efficient planning and production processes.
• Cost predictability, allowing HP to better manage cost uncertainties.
• Assurance of Supply, allowing HP 100% coverage of their demand, even during industry-wide shortages.
• Inventory cost reductions through inventory level optimization both internally and at supplier sites.

In his blog entry "Get real: The Secret of Supply Risk Management" Tim Minahan argues the need to focus on high probability risks rather than on low probability/high impact ones.  In the current economical situation this is particularly the case. The drastic decisions taken by companies will result in shortages down the road as it is extremely difficult at the moment to predict future demand.

Procurement Risk Management allows us to manage uncertainty and adapt our needs at the lowest cost/risk. It was developed for direct material procurement, but has since also been used in indirect procurement. If you are interested in more of the mathematical model, you may want to look at "PRM using Option: Random Spot Price and The Portfolio Effect" by Qi Fu, Chung-Yee Lee and Chung-Piaw Teo from HKUST and NUS Business School.