Saturday, May 14, 2005

Optimization In Supply Chain Management part 2

One area that has always fascinated me is the pace of adoption of optimization in supply chain management. As I had written in a previous article, compared to other industries like airlines and oil and gas trading, the creation of optimization based solutions for supply chain management has lagged. The adoption of these sophisticated tools by manufacturing companies have been even slower. Even leading edge manufacturing companies run their supply chains with planning tools which are very simple heuristics with obvious short comings. Companies seem to be happier with simple understandable solutions (read simple heuristics) rather than sophisticated algorithms. Why is adoption of advanced optimization to supply chain management lagging other industries? Who best to pose this question that to Narayan Venkatasubramanyan. Narayan has been a pioneer and a leader in the application of cutting edge optimization techniques to supply chain problems. In his prior life he had applied optimization to tough problems in the airline industry. In the first part of the interview with Narayan, he talked about the reasons why he thinks adoption happened earlier in Airlines. Here is the second part of that interview.
Like any sweeping generalization, you will find numerous examples to the contrary but I'm inclined to think that airlines have embraced optimization much earlier than manufacturers and hence have reached a higher level of sophistication. There are several distinct reasons for this. We talked about early adoption as one of the reasons in part 1 of the discussion. Here are a few other reasons why airlines have embraced optimization earlier and have reached a higher level of sophisticated when compared to manufacturers:
  1. Dramatic changes hastening acceptance
  2. Embedding in business processes
  3. Early growth of commercially viable companies offering optimization solutions
  4. Customer Relationships
Dramatic changes hastening acceptance
Much of airline optimization began back in the early days of deregulation. This was no accident. Until circa 1980, the airline business was highly regulated. Airlines had little flexibility in terms of what markets they serviced, what prices were charged, etc. The decision to deregulate airlines dramatically changed the choices that airlines faced (Overview of deregulation and its effects). Airlines are large complex beasts. They may be easy to run under fixed rules but when the rules change, the ability of humans to comprehend how best to change in the face of those changing rules is limited. Enter computers, detailed mathematical models, and precise optimization techniques.
Contrast this with the environment for manufacturing industries where there was no such compelling event. It is not as if their environments have been stable, it is just that they have seen constant change but perhaps without changes of the magnitude of deregulation. B2B promised to be one such change but it failed to live up to that expectation. (Karthik's comment: Outsourcing could be such a dramatic change, but it is too early to tell.)
Embedding in business processes
The use of optimization models in airlines became deeply embedded in their operations. Flight scheduling applications did not merely answer the question of fleet composition but were designed to generate detailed schedules that were published for use in reservation systems and flight operations systems. Crew scheduling applications did not simply leave off at solving strategic problems of "how many crew do I need and where?" They took it all the way to operational problems of "who will staff flight AA 123 on Friday, April 29th?". The output of yield management systems were validated and hence trusted to routinely decide when to say "no" even when seats were available.
Contrast this with many of the applications in manufacturing industries where the so-called decision makers do not necessarily have a direct connection to the activity on the shop floor.
Early growth of commercially viable companies offering optimization solutions
In the early days when airlines were beginning to adopt operations research, each airline had its own Operations Research (OR) group. In my opinion, this is not a workable model. OR practitioners are good when it comes to understanding the business problem and turning them into sophisticated mathematical formulations and solving them. There is a lot of infrastructural work, both systems and process, that is essential to make their output meaningful. Until American Airlines Decision Technology (AADT) came along and decided to replicate their successes at American Airlines, there was no mechanism for "learning".
I would argue that until companies like i2 came along, whatever little work was being done in the area of applying optimization to manufacturing had little success because it missed the experience that a "learning" company acquires through its own actions.
Relationship with few big customers
The nature of the markets which airlines and manufacturing companies operate in differ quite widely in one important sense: the former lead with a very large number of customers, each of whom has a very small impact on the airlines strategy; the latter typically has relationships of long standing that cannot be subjected to a myopic optimization model.
Do you agree with Narayan? In the next few weeks, I am hoping to coax him to talk about where he sees the future of optimization in supply chain management.

Karthik Mani


  1. Kathik -- I would look to the boundaries of supply chain processes with other processes for the next big ideas. Optimizations that don't encompass the entire problem are always "suboptimal". Given the continuing trends of even cheaper disk space (free?) and memory -- the 2 trends that kickstarted i2 orgininally, I think companies will be able to take on bigger problems. The challenges as always will be to
    unlock the current data, ensure it is accurate AND someone who can understand what the relevant problems to that business are.

  2. Karthik,

    Interesting I just came across these blogs today.. I am curious to find out your opinion and Narayanan's on why airlines are going bankrupt with all the optimization.. On the other hand I am not sure if Dell does any optimization.. your thoughts...