Of course I am playing off of the title of the movie “The Eternal Sunshine of the Spotless Mind”, a comedy with Jim Carrey and Kate Winslet, in which IMDB describes the plot line as
A couple undergo a procedure to erase each other from their memories when their relationship turns sour, but it is only through the process of loss that they discover what they had to begin with.
And this is the eternal dilemma of decision making too, only here the odd couple is human judgment and machine optimization. In the movie the Jim Carrey character likes consistency and dependency while the Kate Winslet character likes uncertainty and discovery, with the obvious tension from these character differences coming to a boil.
Since the inception of APS (advanced planning systems) in supply chain in the mid-1990s we have focused on getting machines to make optimal solutions and eliminating the person from the decision process. This approach is captured best by the concept that even the smartest people can juggle no more than nine variables when making decisions. Since there are a lot more than nine variables that need to be considered when making a cross-functional decision in supply chain, the solution was to eliminate the people and let a machine make an ‘optimal’ decision.
It is time that we corrected this imbalance.
Whenever we treat one part of the supply chain in isolation, such as forecasting or inventory optimization, it does seem that there is an optimal answer that can be provided by a black box approach. But so many decisions we make in one function have a knock-on effect in other functions, which we seldom take into consideration when we focus on a narrow set of decision criteria applicable to one function.
For example, I know of several cases where optimizing inventory has led to less than optimal use of manufacturing assets. Of course the obvious next step is to ‘rationalize’ the manufacturing assets while focusing a lot of effort on greater manufacturing flexibility. But what if you get caught with a rapid and big shift in mix that has little net effect on revenue or inventory value, but has a big impact on capacity needs?
In addition, all too often we assume in supply chain that we have ‘a’ lead time – a single value – or a yield, or a throughput, or a batch size, or … when in reality all these variables have ranges. (As an aside, that is why they are called a variable.) And then we calculate an ‘optimal’ one-number plan from these approximate input data, generating a result that is ostensibly more accurate than the input data.
We should be focused on a risk adjusted trade-off analysis across the supply network rather than a narrow functionally focused optimization. Of course people need systems to crunch the data. But I do not believe that we are served well by an approach that takes the person out of the loop. We need to couple machine speed with human judgment to make these trade-offs across competing objectives, the relative value of which changes based upon business conditions. For example, typically when revenue is on target margin is the primary focus, but when revenue is lagging, margin is a distant concern. People can handle these nuances and changes in emphasis far better than a machine.
There is a McKinsey article titled “When Toyota met e-commerce: Lean at Amazon” which brings out this dichotomy very well. The paragraph that jumped out at me is below, with my emphasis.
Given the business evolution of Amazon from a bookstore to the store for everything, we had to reinvent automation, following the lean principle of “autonomation”: keep the humans for high-value, complex work and use machines to support those tasks. Humans are extremely creative and flexible. The challenge of course is that sometimes they are tired or angry, and they make mistakes. From a Six Sigma perspective, all humans are considered to be at about a Three Sigma level, meaning that they perform a task with about 93 percent accuracy and 7 percent defects. Autonomation helps human beings perform tasks in a defect-free and safe way by only automating the basic, repetitive, low-value steps in a process. The result is the best of both worlds: a very flexible human being assisted by a machine that brings the process up from Three Sigma to Six Sigma.
Machines should be focused on the ‘Simple’ and part of the ‘Complicated’ quadrants of the Cynefin Framework. This is where scale and efficiency are important and can be realized through the adoption of ‘best practice’ processes. The traditional APS approach even has validity in the Complicated quadrant where the decisions are not that complicated, and can be codified, but a lot of data needs to be analyzed to determine the best path forward. Machines are more efficient than humans with this type of work.
However, we must also not lose sight of the fact that many decisions we need to make in supply chain lie in the ‘Complex’ quadrant. This typically where there are trade-offs to be made between customer service and cost, between capacity utilization and supply chain agility. In short, between efficiency and effectiveness. The answer is not obvious and depends on a multitude of external factors, often out of the immediate control of the team making the decision. The Complex quadrant is where we see every day disruptions in our supply chains which have a significant financial and operational impact; often due to several disruptions have a multiplying effect. The Chaotic quadrant is where we see catastrophic disruptions such as earthquakes, floods, etc. While many companies have disaster recovery plans, determining the exact response to a specific incident often requires trial and error, something at which a machine is not very good, especially when the exact conditions have not been experienced before.
An interesting aspect of the Cynefin Framework is that the Simple and Chaotic quadrants are next to each other. This represents the fact that when we focus too much on efficiency, processes in fact become rigid and fragile. They are unable to absorb even small variations and disruptions. As importantly, the center shaded area is Disorder where we do not understand in which quadrant we are operating, the worst of all possible worlds. There is a great short introductory video to the Cynefin Framework that explains this is greater detail from about 6:30 minutes. The McKinsey article on Amazon’s use of Lean principles also captures the delicate balance between not over simplifying a process so that it can be automated, and therefore provides scale, and given people the flexibility to apply human judgment to situations, and therefore provide flexibility and agility.
I can only repeat: It is time that we corrected the imbalance between machine-based optimization and human-based judgment.