Lean Methodology and Waste in Data Processing
“Just-in-time” is a phrase currently in the news in the context of potential disruption arising from leaving the European Union. It is sometimes portrayed as an unnecessary exposure to the risk of an interruption in supply. In fact, “lean” methodology can accommodate contingency planning for supply risks; it also helps you to calculate how much the contingency arrangements add to cost.
Lean methodology was developed in Japan in the years following the Second World War, when extreme efficiency was a necessity, not a choice, because, amid the destruction, resources were extremely scarce. Lean methodology began in manufacturing, but also applies to data processing, which is what this series of blogs will explore.
What Does Lean Mean?
Briefly, the principles of lean are these:
- Value Identification: it is essential to determine what is the value that is being delivered by a process. Otherwise there is no effective basis for assessing the process.
- Mapping, which demonstrates how every activity fits in, and how it adds value.
- Flow, where processes are synchronised so that there are no bottlenecks or build-ups of inventory or work in progress.
- Continuous Improvement, so that the available resources generate increasing value.
- Pull, where production is driven by customer demand, and varies precisely in line with it.
In a lean organisation every process is devoted to adding value for the customer, and every employee understands how he or she fits into that process. Each employee knows that delivering his or her contribution right and at the right time contributes to what the customer requires. And, on the other hand, no employee is spending time on unmapped activities.
Don’t get pushed about. Pull your information.
In data processing the value consists in the opportunities and shortcomings that are brought to light by the information generated. Mapping, flow and continuous improvement are all also clearly applicable in this context. “Pull”, though, implies:
“Don’t generate a report just because it is Tuesday, or just because it is the third day of the month. Instead generate it whenever, but only when, there is something interesting or important to say.”
In other words, there is no point reporting in great detail that there is nothing to report. Indeed, it is wasteful to do so.
Hence the concept of the dashboard. In its purest form, a business intelligence dashboard would consist of maybe just a couple of visualisations that are always on, and a host of others that only appear when they are of current interest, but then tell the user all he or she needs in order to act on the information.
Eight Deadly Wastes
The practices that follow from the lean principles include the identification and elimination of waste, in eight categories:
- Transportation and transmission of resources and the finished product
- Inventory of resources, work in progress and the finished product
- Motion of the person carrying out the process step
- Waiting for the resources needed to carry out a step, or to deliver to the next person in the process
- Over-processing: spending time and other resources on features that the customer does not require
- Overproduction: producing more than the customers require
- Not utilising human resources, and particularly neglecting to solicit employees’ suggestions for improvements
Waste reduction is never complete. There is always room for improvement. So this is where continuous improvement comes in; this is about making incremental changes to reduce waste and improve efficiency.
In this series of blogs I will take each of the wastes in turn, and consider how it applies to the processes of data processing.