Let’s use an example, let’s say I’ve been asked to increase the efficiency of a cell in a manufacturing plant as it’s currently loss making. I’ve been informed that its costs are mainly associated with manning and the amount of hours it takes to complete a product. The company requires a saving of at least 50% in order to make a profit.
The first thing to do is get an understanding of how the product is manufactured. If you have planned operations available to you then that’s great make full use of them. I find the best thing to do with this is create a standard process map so you can visualize the steps in manufacture. If there’s no planning available then the likelihood is it’s a real simple process and you can map the process just by observing it.
If you’re lucky the company you’re working for may have standard times against each process. These are usually created by industrial engineers or the planners taking advantage of a timing tree created by an Industrial engineering specialist. These standard times can be used as a reference against your actual times to complete the processes in order to determine the level of performance. Why this is important? Well if your performances aren’t around 75% or more you know there’s definitely cost savings to be made via improving performance of individuals. In fact if the performances are really low say 30% then you immediately know there’s plenty of fat on the bone to be attacking and get those costs down. On the other hand let’s say I need to reduce cost by 50% and the performances are currently all at 60% against my standard times. A world class performance rating is often said to be 82% but this can vary based upon the complexity of the task. So if 82% is deemed world class then 82% – 60% = 22% .This means If we can get the performance levels up to world class level I have a potential gain of 22%.I needed to gain 50% to make the product profitable to the company I’m working for. This tells me straight away that increasing performance alone through efficiency gains is simply not going to cut the mustard, I’m going to have to look for other opportunities. If the performance gains available can get you to your desired saving then at this point you need to start studying the wastes. Standard times assume there are none of the 7 wastes effecting your manufacture so any you can find and address will lead to performance gains hence reduction in cost on the cell. But for now we just want to find opportunities and brief our leadership on them so let’s move on.
If you don’t have standard times against operations available to you then your best option is to observe the processes. Get a feel for how efficient and how hard you think people are working and use that as a gauge later on.
Ok so my cell has performance gains available of potentially 22%. That leaves me with another 26% to find.
The 7 wastes are otherwise known as TIMWOOD. I’m not going to go into too much detail at this point but these stand for
Transport, Inventory, Motion, Waiting, Over production, Over processing and Defects.
Of these wastes only, Motion, Transport and Waiting really have an effect on the performance of my cell operators and therefore their performance levels. I can target these with 5s exercises but we will talk about that on another occasion, for now let’s just stick to a systematic approach to gain an understanding of the manufacturing cell and what improvement oppertunities
The major waste we need to look at now is Defects. To look at this we need to find our production yields. We need to get the yields at each stage of manufacture. If you’re lucky you will have this data readily available to you via one of your engineering support teams. If you haven’t got this information readily available you can either try and collate it or use a value stream mapping exercise and ask stakeholders for a rough idea. You can then use this data to calculate the rolling yield. The rolling yield is the total yield for the cell.
At this point I would be assessing the individual yields. Are there any that are particularly low? Again like the performance levels if the yields are particularly low, let’s say 50% then that means there’s potentially plenty of meat on the bone to go at. If the yields are pretty good or perfect then you know this isn’t an avenue to spend too much time on as you’re only going to find marginal gains.
From my experience the best way to find opportunities to increase yield is through improvement events where both the support staff and operators are invited to feedback on ideas they have to improve quality. I promise to write a blog ad chuck in some templates on how I do this sometime in the near future. A more formal way of finding solutions is to use a practical problem solving exercise with the intention of getting to all the route causes of the issues and implementing solutions. Again I will discuss this in a later blog.
What’s next? The next thing id look at is utilization of resource. You still have to pay for resource if it’s working or not. In other words it’s costing the company for a person or machine to sit around doing nothing whilst it waits for work. You can work this out in a number of ways but the easiest is to use historic data.
Resource worked hrs vs Available hours.
Let’s say this works out at 50% that means with some effective work balancing we could expect to get that up to at least 90% meaning a gain of 40% in utilization this means we can either increase production rate and reduce cost against each product or manufacture the same quantity with less resource hence less cost.
We have found that a 22% increase of performance of the cell operators can be achieved via targeting waste and managing the operators effectively with key performance indicators.
The industry standard for yield in this manufacturing field is 70% in fact we have achieved yields of 80% ourselves in recent history. However our current yield is just 50%. As such it can be expected that we could achieve yield gains of at least 20% through improvement activities.
Our current utilization of resource is averaging 50%. We believe that for this product utilization of 90% can be achieved meaning a potential gain of 40%