Process Engineering: rebalancing a mechanical assembly line to increase efficiency and productivity

Saline Lectronics Blog

As discussed in previous posts, Saline Lectronics process engineers are the magicians behind the curtains actualizing the ideal manufacturing process for each circuit board assembly that we build. Our process engineers are responsible for optimizing manufacturing operations to help the assembly process flow as smoothly as possible.

As process engineers, our team is responsible to “de-bottleneck” areas – in other words, work out any process kinks that prevent the operation from functioning properly or efficiently. Most recently, our process engineering team performed a thorough time and motion study on the mechanical assembly line of a high volume product in order to properly balance the line, remove waste, increase efficiency and reduce the overall cycle time.

mechanical assembly line

Mechanical assembly flow line at Saline Lectronics.

This specific mechanical assembly line was originally designed to have seven different workstations with an average cycle time of 75 seconds per station and a daily production goal of 300 completed units per day.  Over time, the process engineering group noticed that the line was consistently producing above target with an average cycle time of 60-65 seconds and an average throughput of 350+ completed units per day.

While it may sound positive to continuously exceed set targets, it actually triggers a red flag to our process engineers that a line needs to be examined and most likely, re-balanced. It’s also important that targets are realistic so the scheduling team can appropriately plan the daily production agenda.

During the time and motion study, process engineer, Denise Kuenzer, needed to establish a baseline of actual performance. With a stop watch in hand, she monitored and timed the entire assembly line ­­– she looked at the time to perform the work at each individual station as well as the the time it took to transfer work from one station to another. Once the current process was established, Kuenzer was able to properly analyze the gathered data and identify inefficiencies. In her study, she found unexpected waste in assembly work that was split between operations, and waste in the time lost to transfer a part from one station to the next.

In the case of unevenly split assembly work, one station experienced a much shorter cycle time than the other stations. The operator at this station was typically waiting around for a part because it took far less time to complete the required operations. To better balance the workflow, the process engineering team rearranged specific assembly steps to be completed at different stations. For example, instead of placing the label at one station and actually sticking it to the product at the next station, the entire assembly work of affixing the label to the product is now performed at one station.

Re-balancing the work on a lean flow line can take time. Additional time and motion studies need to be performed to verify that the adjusted workflow is conducive to an even and efficient line. Before updating the mechanical assembly work instructions, our process engineering team validated that these changes actually evened out the assembly flow and kept the line moving smoothly.

On this mechanical assembly line, Kuenzer also observed that there was a lot of time wasted in parts transferring from one station to the next. For example, she timed that it took 13 seconds for an operator at station four to walk a part to the next station. To address this lost time, she simply adjusted the distance between the two stations. During Kuenzer’s study, she also noticed the automatic conveyer belt moving product experienced a 10 second delay before it actually moving. Initially this delay perplexed our process engineering team, but after digging into it further, they discovered that the software code originally written to trigger the movement of the conveyer belt was losing time because it was storing too much excess data. Process engineer, Jeff Eisemann, re-wrote the code and now the conveyer belt initiates movement two seconds after confirmation.

Since making these small changes to this mechanical assembly line, it is now functioning at optimal performance. It’s a six station flowline with even workflow and an average cycle time of 55 seconds. With one less operator, the line consistently produces 350 to 400 finished units per day.

The work performed by our process engineering group is in a constant state of check and adjust from the PDCA cycle. Improvements are delivered, analyzed, measured, re-analyzed, adjusted, re-analyzed, etc. We’re always monitoring what we’re doing. Interestingly, these gains in efficiency didn’t require a single person to work harder or faster; it’s simply about looking at the process and making it smarter.