Industrial productivity: how to deliver more from assets?
Industrial productivity is a key challenge. Productivity growth from existing assets has several positive impacts on profitability:
- Reduced fixed cost per ton or per product.
- Increased revenue.
- Reduced capital expenditure as capacity investments can be postponed.
When manufacturing sector plants think about productivity, they systematically consider OEE measurement. It is a good starting point as it gives a standardize approach to measure productivity, and first insights about which type of issues are causing lower productivity. However, it will give you limited insights to really act of labor productivity. The reason is that OEE is only the thermometer measuring the consequences many aspects have on productivity. It lacks the story about the root causes.
“Productivity is not just about OEE”
How to improve productivity? This can be done in two complementary ways. First by tackling all the loss of productivity (Muda) caused by abnormal production process behaviour. Production line stoppage, product scraps… The second approach is to learn from the production process and find new ways to operate at a higher pace. In both cases, OEE will not be sufficient.
Here comes data. Data is a key lever to act on productivity. If you bring together data coming from OEE (stoppage time, its cause, production and scraps volumes…) with manufacturing process data (process parameters, raw material data, quality controls, equipment cycles details…) you will simplify the work of your team of experts for both troubleshooting and improvement actions.
With the proper information and tools, you can:
- identify the root causes of your production stoppage and tackle them
- reduce your process variability and avoid out-of-spec products
- optimize your production line equipment usage (reducing waiting time, unit operation time, increasing absolute throughput…)
- reduce equipment’s down time with conditional and predictive maintenance.
- get insights to rethink your organisation
For instance, in many manufacturing processes you can find some dosing steps. Let’s consider a process where it is necessary to dose several ingredients. By collecting each ingredient dosing duration, you can identify which ones are necessitating the longer time. Then, you can investigate variability of theses dosing times and find ways to reduce them and reduce their variability. At the end of the day, you have increased the throughput of you dosing equipment. This is just one of many examples.
Data is an enabler, and it must be combined with proper methodology (root cause analyses…) and business expertise from your teams. Altogether, you can expect not only to avoid many annoying incidents impacting your productivity, but also find innovative ways to operate and increase your assets throughput.
Mathieu CURA, January 2022