Over the last 30 years, I have evaluated the operations of hundreds of manufacturing, processing, mining, and service organizations. Three things I have learned from that work:
- Companies know their business, but their knowledge is incomplete. They shoot targets, monitor KPIs, and compare their performance to their plans and budgets. But they don’t have time to actually think about their operations and performance.
- Companies have lots of data. It is not unusual to ask for real data and have a management team laugh heartily and say: “Oh, we have tons of data. So much data that you will be drowned in data.”
- Companies do not look at their data very much. Unfortunately, when asked what their mountain of data tells them about their operations, they go back to their targets and KPIs. They have data, but they use very little of that data because they believe there is too much data to understand what it is telling them.
- This lack of data analysis represents lost opportunities to compete more effectively within their industry. Some quick examples:
- A world-wide, leading mining company was experiencing lower productivity on weekdays than weekends, resulting in more than a 7% (overall) loss in production capacity. Approximately 2% of this lost production was due to necessary operational and maintenance downtime, and the remaining 5% was unexplained. Why was it unexplained? The majority of lost production mid-week was attributed to the planned downtime. Since the available operational data had not been analyzed, there was no way to know that there was a potential to increase output by 5% by understanding and correcting the weekday operating problems. Instead, the problems were seen as insignificant. Imagine the value of improving an operation’s performance by 5% just by implementing procedures that are already in use within the organization.
- An electronics manufacturer was experiencing over 25% lost capacity, resulting from shutdowns at lunches and breaks, on a board assembly line. In addition and for unexplained reasons, the line productivity ramped up as a day progressed, peaking mid-day on day-shift, and then fell off again as the remainder of the day was completed. This ramp-up/down phenomenon resulted in a 40% additional loss in capacity. In other words, the line capacity could have increased by an additional 65% just by operating continuously at its demonstrated peak productivity level yet no one knew the opportunity existed. Once again, imagine the impact of increasing capacity by 65% without additional investment This may push new capital investment off by years, not to mention the improved labor cost per unit impacts.
In this age of computers, these types of opportunities seem surprising. Yet it is not unusual to run into a management team that does what I sometimes think of as “managing by cocktail napkin”Β. Each day they face a set of problems indicated by their performance tracking measures but not really understood. They discuss possible solutions over lunch, take some notes on their cocktail napkins, finish lunch, and throw the napkins away because they still don’t understand the cause of the problems. The next day, they go through the same process all over again, never once looking at the underlying data from their operations.
With some effort and the willingness to scale a mountain of data (hundreds of thousands of data points in the cases above), the opportunities are clear. The data is available, but time for the analysis is not invested. The magnitude of these opportunities suggests that investing in this analysis will often result in a very real and fast payoff.
The bottom line here is that companies that do this type of analysis gain a competitive advantage in their industries. Since we do some of this analysis as part of our shift schedule evaluation projects, our clients benefit when they evaluate schedule alternatives. But you don’t need to be changing your schedule to do the analysis. You do need to know how to do the analysis or hire someone that knows how to do the analysis. And you should do that soon because your success is at stake.
I will be writing more about this topic over the coming months, including:
- Looking at some example data sets and sources.
- What questions should be asked of the data?
- Answering these questions using some of the analysis tools already on your computer.
Of course, we are happy to help you if you would like outside help. Call or text us today at (415) 858-8585.