Most small manufacturers simply can’t afford to have an ERP system that tracks manufacturing cycle times and the variances between those times. However, even for those companies with the most up-to-date software, there are some inherent benefits of witnessing production happen in person. In fact, even the best software isn’t intuitive enough to show you how to eliminate idle time and increase production throughput. For that, you have to see work being done for yourself. With this in mind, I thought I would include a cycle time tracking excel sheet for small manufacturers with a built in graph that shows average, median and mode cycle times in a given production work station.
Granted, excel isn’t live and ultimately isn’t the best tool to track production. However, it’s a fantastic tool to establish your benchmark cycle times in manufacturing. It’s these ideal cycle times that should be used to measure all future variances. Why? Well, one of the biggest issues I encounter is when one of my manufacturing customers takes the times emerging from their MRP systems as gospel. I always hear “yeah, we know what our cycles times are from our system” to which I always reply with “that’s great, but do you know what they should be?”
As mentioned, no system is intuitive enough to show you how to reduce your production times. All any system can do is track these times and show you variances. In order to reduce waste in manufacturing, you really have to witness production happen with your own eyes. In this regard, using an excel sheet to establish your ideal cycle time is the best way to measure future variances in or out of your ERP system – if you’re lucky enough to have one.
You’ll better understand why some times are lower than others and ultimately why a particular work station experiences higher than normal cycle times. To succeed, you need to use the five steps I outlined in the post Manufacturing Cycle Times in the Perfect Work Cell.
Production Work Station Analysis
The following table outlines the cycle time analysis that took place at a given customer’s account. There were a total of twenty sample cycle times (broken BOLD BLUE line on the graph) taken during a single shift. This particular table outlines one of our initial trials. The twenty times were placed on a graph in order to establish the natural progression of times from each work task. We established an average (“mean”) time, isolated the median and the mode of the entire 20 samples. Keep in mind, these were 20 consecutive samples taken over a single shift. We then replicated this exercise every day to isolate the causes of lost time.
Average (“mean”) Times: Calculating the average simply involves totaling up all the twenty times and dividing it by the number of operations. In this case, it’s 51 minutes divided by 20 operations. This gives us an average cycle time of 2.55 (two minutes & 55 seconds). The AQUA BLUE straight line represents the average.
Median Time: To calculate the median, simply rewrite the entire sequence of times in order. Next, use the answer form the calculation below to isolate the median time. While I didn't include the median in the graph, you could. It's simply another measurement of average.
- Median: {(n+1) / 2}
- N= sample size: which in our case is 20 operations
- Median: {(20+1) / 2} = 10.5
2.15, 2.15, 2.25, 2.3, 2.3, 2.3, 2.3, 2.3, 2.45, 2.45, 2.45, 2.5, 2.5, 2.5, 2.5, 2.55, 3.25, 3.25, 3.25, 3.3
Mode time: This is simply the time that occurs the most often in the sequence. In our example, the most common cycle time is 2.3 (two minutes & 30 seconds) The mode time is RED straight line on the graph.
The Analysis of the Graph
Don’t be too concerned about the differences between mean, median and mode averages. The mode points to the most consistent time. In essence, you could argue that this is your ideal cycle time or benchmark time to measure all future variances off of. Your average time can be skewed by a few higher than normal results – such as the four times that were above 3 minutes (3.25, 3.25, 3.25, 3.3). The median (which is not depicted on our graph) simply separates the top half of the sample from the bottom half of the sample.
The purpose of the exercise is to identify your highest cycle times and answer the following three questions.
1. Why are these cycle times high?
Identify the causes of the high cycle times. Causes might include incomplete bill of materials, an inaccurate work order, a confusing assembly outline, or any and all issues pertaining to equipment usage.
2. How often does this happen in work shift, day, week & month?
You must quantify the costs of idle time in manufacturing and establish the company’s costs over a given shift, through all the shifts within that day and summarize the impacts over an entire week and month. It’s therefore important to analyze results over several days, in order to have a more accurate depiction of lost time.
3. How can this be fixed?
Identifying the causes of interuptions is often much easier than finding a solution. Those aforementioned issues pertaining to work instructions are fairly easy to resolve. They take time to fix, but are often well worth the effort. Issues pertaining to machine repairs are something else entirely. For this to work, you must do a cost-benefit analysis on the cost of repairs versus the costs to the company of not fixing these interuptions.
While it’s not an exact step-by-step process to performing a cost-benefit analysis, the following post still provides some insight into making a decision on an equipment upgrade. What's Involved in Making a Decision on a Capital Expenditure?
Posts like this brigehtn up my day. Thanks for taking the time.
Posted by: Yegor | 02/13/2012 at 05:41 AM