Part of global industrial conglomerate Siemens AG, Lincoln-based Siemens Industrial Turbomachinery designs and manufactures industrial gas turbines, exporting over 80% of production to a worldwide customer base.
“By the time that production issues were flagged in our ERP system, operators had moved on to other tasks, making problem identification difficult.”
Although built around common cores, each industrial gas turbine project is unique, typically involving an order of between one and twenty gas turbines, each configured to customer requirements with one hundred or so major sub-assemblies, and a bespoke turbine control system.
This particular plant’s assembly operation is capable of working on ten industrial gas turbines at once, along with orders of spare parts and sub-assemblies. A typical customer order, notes manufacturing operations manager Adrian Toyne, equates to more than 1,000 standard hours of work per unit, although spare parts orders can consume as little as 15 minutes.
The problem faced by the plant? Principally, explains Toyne, a persistent gap between targeted efficiencies and actual achieved efficiencies.
“We’d spent two years making improvements in our operations, trying to bring our efficiency levels closer to the target,” he relates. “Everyone appeared to be hard at work, and there were no obvious flaws in the standard times. Yet still the efficiency gap remained.”
Moreover, he adds, there were issues with how the plant recorded those actual hours that had been consumed. Operators’ times on jobs were first of all manually entered onto timesheets, and then subsequently typed into the plant’s SAP ERP system.
Worse, this manual booking of actual hours into the SAP system provided insufficient granularity into production tasks and problems. As a result of its focus on charging work content to individual customer orders, the system took a costing-centric view of hours worked, as opposed to providing information which would be helpful to factory-floor management pursuing a continuous improvement agenda.
In other words, says Toyne, when operators were unproductive, it was difficult to discover why, so that preventative measures could be put in place to avoid a reoccurrence.
“The lack of granularity, coupled to the time delay, made it difficult to see the cause of any problems,” says Toyne. “By the time that production problems were flagged in the ERP system, operators had moved on to other tasks, making problem identification difficult. In a business which had begun its lean manufacturing journey, we had critical gaps in our basic understanding of what was going on.”
At which point, he relates, Siemens’ plant management read about the Mestec manufacturing system in The Manufacturer magazine.
So might Mestec’s factory-floor data recording terminals be a way of closing the gap? Toyne set about finding out.
Talking to Mestec, a solution very quickly emerged. A pilot using Mestec’s hosted SaaS solution allowed Siemens to try Mestec with no demands placed on Siemen’s busy IT department. The pilot was able to quickly demonstrate the value of the Mestec manufacturing system in helping Siemens to tackle the issues that it was facing.
“Increasingly, our sales model involves customers trying out the Mestec system for themselves, leveraging the ease with which our hosting platform can accommodate new users,” explains Mestec managing director Jeremy Harford. “Not only does this tend to be faster than a traditional sales cycle in terms of customers ultimately deciding to move to a Mestec manufacturing system, but the results demonstrated by a pilot project typically provide an added impetus to full plant-wide implementation.”
Accordingly, in July 2013, the pilot project began.
Very quickly, some surprising answers became evident. Straightaway, for instance, it was clear that the plant’s SAP system and the Mestec system were producing different efficiency figures—with the Mestec figures being closer to what plant management believed was the real situation.
“We discovered that the definition of ‘good hours’ in our SAP system excluded set-up times,” says Toyne. “That single discrepancy accounted for seven percentage points of the difference that we were seeing—and no one knew. Yet we’d been beating ourselves up over it for two years.”
Another significant finding: despite operators’ belief that works order standard times were more or less correct, there was indeed a degree of underestimate, typically of around 11%.
“Everyone appeared to be hard at work, and there were no obvious flaws in the standard times. Yet still the efficiency gap remained.”
And again, the Mestec system proved its worth in providing detailed analysis of non-value adding time. The time taken in rectifying non-conforming items received from suppliers, for instance, had been estimated to account for 5% of actual hours.
But in fact, the real figure was closer to 10%—and being able to flag it in real time, rather than on paper systems, meant that it was much easier to then recover the cost from the external suppliers involved.
The combined effect of all these findings, sums up Toyne, was to rapidly close the gap between targeted efficiency and actual efficiency. Very quickly, Siemens management realised that there was a solid business case for extending the pilot right across the operations area—and in the process, formally integrating Mestec with the plant’s IT infrastructure including SAP.
As a result, in May 2014, Siemens Industrial Turbomachinery went fully live with Mestec’s manufacturing system. The plant elected to remain on Mestec’s award-winning hosting platform, sums up Toyne, but dispensed with 3G wireless connectivity in favour of a fully hard-wired connection to Mestec’s servers.