Unsolved Problems

I have a certain fondness for reading about mathematics. One topic that I find especially interesting concerns the “great unsolved problems” in mathematics. A good example is Fermat’s Last Theorem. Fermat conjectured that there are no whole-number solutions for the equation

an + bn = cn,

where a, b, c, and n are all integers.[1] I don’t lay awake at night thinking about such problems, but it is interesting to read about the various strategies mathematicians have attempted as they struggled to solve these problems.

I think there are also a handful of “great unsolved problems” in Supply Chain process management which are very simple to state, painfully difficult to solve, and, if completely solved, would radically change the entire field. One great unsolved process problem is “process discovery.” I believe the biggest stumbling block that many companies face as they seek to become more supply chain process-centric is the difficulty of discovering and defining their existing processes. Process discovery is tedious and expensive. Unfortunately, efficient supply chain management is entirely predicated on defining a company’s existing business processes. Without a significant initial investment in process analysis, it is very, very difficult to make real progress in process management. In some cases, process analysis takes so long that, once you begin to make improvements the process has already “spontaneously changed.” In other cases, teams struggle to define a process, and then decide that the process is so expensive, relative to the gains, that the organization should halt further investment in the process.

If we could find a relatively inexpensive, efficient way to capture and analyze a company’s supply chain business processes, companies would be freed to do more sophisticated things with their processes. I believe we are on the verge of solving this problem. I predict that we will soon develop an efficient approach based on one of two possible strategies. One possible strategy is that we will perfect methods that allow us to capture processes by monitoring internal business data networks.

The other is that some combination of monitoring, ERP, and workflow will result in an approach that allows large improvements in efficient process capture. Either would enable a great many more businesses to start focusing on supply chain logistics networks and process systematically.

The second great Supply Chain problem is “process measurement.” We need standards that define a nomenclature for common business processes and we need standard ways of measuring process results. Without widespread standards each company, division, or department that undertakes process work becomes an island unto itself. Without standards benchmarking isn’t very efficient or useful.

Similarly, inter-company linkages are very difficult to establish without common names or standard measures that all parties can rely on.

Again, a solution seems to be on the horizon. Standard process frameworks, like the Supply Chain Council’s SCOR framework, define common process names and measures. SCOR has enabled a number of benchmarking companies to provide very accurate, industry specific data on supply chain processes. The widespread adoption of SCOR, and the development of similar business framework initiatives for other common business process domains should begin to provide the standard nomenclature and the measures we need in the near future.

The third problem is “process organization.” How do you actually organize a supply chain process-centric company? Today’s companies already have a financial organization with cost centers, charts of accounts, and so on. Most companies also have a functional organization with divisions and departments. Many companies also have product-line organizations. Now we are asking executives to add “process” to the mix. It is absolutely critical to process performance that the process has an “owner,” that the process owner’s performance objectives are linked to the process performance objectives, and so on. The problem, however, is to figure out how is integrate a process organization with existing financial, functional and product line organizations and to determine how to evaluate the performance of each. I still haven’t seen the right combination of process ownership, organization structure, and business performance management.

I hope that in the coming decade, there is a technological solution to supply chain process discovery, that there is a “governance” solution to process measurement and that there is a wonderful management solution to process organization. In the meanwhile, we just keep “slogging away” and make incremental improvements to the overall process landscape. As I suggested, these are hard problems.

[1] With the help of computers, Fermat’s conjecture has been proven to be true for all possible combinations of a, b, and c, for values of n up to about 100,000, and most mathematicians now believe the theorem to be true. Some still wonder, however, if Fermat had a specific proof in mind that has eluded researchers since Fermat’s death.