Ava’s team is notified of a slightly higher number of returns coming in from distributors.
The next day, Ava checks her MES system but it has no alerts given nothing is out of bounds but the rejections continue.
Pashi lets you view real-time production metrics and create KPIs customized to your industry and process.
Having looked at Pashi's analytics view the previous night, Ava already knows that the rejections are not unique to any particular SKU, though one of the variants is more frequently represented among those rejected. Graphing rejects against time, she notices that the rejections appear to happen in bunches, always after an initial run of the frequently represented SKU. Ava decides to install temporary QA stages after each stage of the process. Since she doesn’t want to unnecessarily slow down the line, she only activates those stages for the SKU that appears to initially trigger the problem. Using Pashi, she is able to quickly design check sheets for each QA stage and deploy them using inexpensive Android tablets. Setting this up takes less than a day and slows the line down by about 2%.
Ava is presented charts prepared by her team after they manually observe many of the cycles over a few days but the reports still show “OK” and the team is puzzled. Ava decides to observe any potential variation for a few more days. Meanwhile, Ava is informed that sales are down by 5%.
Pashi lets you create dynamic, customizable human interfaces including QA checksheets, operator instructions, and rich data input interfaces, using a blazing fast visual UI integrated into Pashi's production flow editor. These interfaces can be deployed in real-time to any network-connected interface devices like tablets, TVs, smartphones or line displays within a production facility.
After two days of running the production line with the newly added QA stages, Ava and the QA team find a spike in rejections after the milling process. Since the offending SKU requires tool changeover, they theorize that the operators are not keeping track of tool changeover cycle counts, thus not changing worn out tools at the specified intervals. They suspect that this is leading to micro-cracks in the tool and improperly milled slots in the product. Ava quickly adds an Android tablet to the milling stage with an input interface designed and deployed in Pashi. In this interface operators can enter the exact times at which the tool was changed during their shift, and are reminded when a tool’s cycle count has exceeded a specified limit. Meanwhile, Ava is infomed by Tom, the plant director, that sales are down by 5%.
Large numbers of consumer complaints suggest that some units of the product are structurally failing far earlier than they should. Ava asks her team to investigate. A SKU-based categorization of rejects indicates that while one SKU has a particularly high reject rate, a number of them have abnormally high reject rates compared to the past. Production engineers chalk this down to human error during assembly due to employee turnover and seek to give production line operators additional training to compensate. Sales are now down by 10%.
Pashi allows you to monitor the execution of process logic and all variable values live when the line is running, enabling fast and effective debugging.
A few days later with the tool changeover interface running in Pashi, Ava realizes that though the operators are changing the tools at the specified intervals, the quality check still shows improperly machined products. Ava deep dives by inspecting the process parameters of the milling machine in real-time using the Pashi Live View while it operates on the offending SKU. She notices that the tool temperature is over the upper threshold limit during the latter part of the milling operation. This causes the cutting edge to deform, an effect which lasts for several subsequent cycles. This deformed edge produces a defective product. Ava reports this problem to the product team and begins brainstorming potential solutions. Sales are now down by 10%.
Consumer complaints are much higher, causing AceTech’s market share to fall due to negative publicity. Despite retraining, abnormally high reject rates for some SKUs persist. Ava decides to add additional QA stages after each stage of the production process in order to find the source of the error. But for this Ava has to order custom HMI devices that will take some weeks to arrive. In the meantime, she decides to set up temporary new QA stages which use manual data entry via pen and paper.
Pashi lets production engineers change parameters quickly using the visual interface while also ensuring safety by allowing reversion to the previous parameters across the line with a single click.
Ava decides to slow down the feed rate and cutting speed of the milling machine to solve the tool deformation issue. They test out the line with the machine slowed down for a few days and observe the trend in the Pashi analytics tab, which automatically appends data and updates in real time.
Ava’s team is collecting data using manual methods leading to messy and incomplete information. Ava’s team attempts to consolidate the data in a spreadsheet, making a graph of average rejection rate at each stage. Their data suggests that the issue is happening at a manual assembly stage right after the milling stage. The team places the most experienced operators at this stage for a few days to try and avoid error.
Pashi lets you quickly deploy and connect temporary stages to deal with short term issues and allows for quick disengagement of those stages when the issues are resolved.
Ava finds that the reject rate is back to normal. This solution however increases cycle time and reduces production line throughput by 50%, which is unacceptable. Ava decides to compensate for the increased cycle time by installing a spare milling machine from an inactive production line to the workflow, which will run in parallel to the other milling stage. She also suggests changes to the product design that would lessen the tool load during the milling process.
Not seeing any improvement with the experienced operators thus far, Ava’s team is at a loss. Ava calls on maintenance technicians to have a closer look by dismantling the machine. The maintenance technician informs Ava that his maintenance logs don’t show operators performing regular milling tool changes required for the second SKU and thinks it could be a reason for the defects. Ava calls the SCADA system contractor to configure the milling machine for process data capture.
Pashi allows for a customized and complex production routing including splitting and merging production lines, non-linear production flow and programmable flow logic.
Ava has installed the spare milling machine on the production line and quickly writes flow logic in Pashi to split the flow of products between the two milling machines and then merge the output of both machines back into the normal flow of the line. Ava runs the modified production line.
The HMI devices finally arrive but they need to be configured and programmed before they are functional on the production line. Ava asks the HMI supplier to add a tool change input interface in addition to the QA checksheets to confirm the maintenance technician's theory. This causes delays as the supplier requires a change request and re-programming of the HMI logic.
Ava confirms that the product throughput is now back to normal and based on the data from the temporary QA stages, the product has zero quality issues. She then uses Pashi’s genealogy feature to trace previously shipped units affected by the abnormalities in the parameters. She sends this report to the Quality group.
The external contractor reconfigures the line’s SCADA system to capture process parameters from the milling machine. The analysis of a day’s data from the machine confirms the maintenance engineer’s theory. Ava realizes that the tool temperature is above the upper limit during the latter half of the milling process for the SKU with the most failures, causing the tool to deform. This compromises the structural integrity of the product and affects subsequent milling cycles for other SKUs. Ava decides to drastically reduce the feed rate and cutting rate of the milling process, and verifies that products now come out of the milling machine without structural damage. Ava lets the product team know about the issue. The Quality team asks Ava to provide her data to identify units that need to be recalled.
The recall and replacement of the faulty units identified by Ava is complete. AceTech’s customers are satisfied with the speed of redressal. Sales have climbed back and are now down only by 7.5% as the recall is executed quickly.
Since Ava cannot trace process parameters to individual units, she is unable to identify specific units affected by the abnormalities of the milling process parameters. Due to this, AceTech initiates a wide-ranging recall of all units that were produced in a production run with the offending SKU. Sales are now down by 15%.
Sales are now at normal levels for AceTech. The product team has come back with minor changes to the design of a few parts that ensure that the load on the tool during the milling process is reduced. Ava makes changes to the Pashi Program such that the new product design is produced on one of the milling machines while the original is produced at a slower speed on the other.
AceTech’s recall is still ongoing due to its sheer volume and some of the production force is diverted to help disassemble the returning faulty units. The line is now running at half speed production to support the recall. The product team has come up with minor changes to the design that require only a few parts that ensure that the load on the tool during the milling process is reduced. This however, requires external contractors to change the PLC logic for the affected stages. Meanwhile, Tom, the plant director, is now involved as sales are down by 20%.
Easily remove temporary stages and re-establish previous connections.
After three days of production, it is clear that the new product design does not have any structural flaws after the milling process so Ava deactivates the spare milling machine and puts it back on its original production line. In just a few clicks, Ava also disables the temporary QA stages that were placed at various points on the line to increase throughput further.
After one month of half-speed production, AceTech has finally completed the recall. The PLC logic for the affected stages has been rewritten and the line is now able to run at optimal throughput without rejection rate issues. The offending SKUs are now passing quality checks. Ava can now dedicate labor for the manufacture of replacement units.
AceTech reverses the minor dip in its share prices and brand equity and its revenues return back. on track.
AceTech has finished shipping out all replacement units. Sales are now down only by 5% but it's small consolation as AceTech has lost signigicant brand equity and market share due to the product failures and recalls.
AceTech’s demand for the product increases and they prepare for an expansion.