Teams of people in almost every company are focused on expanding and improving the utilization of data from the manufacturing and shop floors. While multiple benefits accrue from each new successful initiative, including part traceability records and process trending. The driving imperative is a quality improvement. We hear about autonomous driving vehicles, yet almost all of us get behind the wheel each day to get to where we need to go.
Relevant buzz words such as “ The Internet of Things (IOT)” and its formal mate “Industry 4.0” for machine communication and monitoring sound cutting edge for today’s heat treat environment. It is an abundant data age brought on by Tera Drives and Quad-Core Processors. Just like a supercomputer’s Gigabyte, digital capability delivers more reliable results faster, and it seems logical that a smart furnace or a single piece flow Induction machine, built with real-time data capture, should be able to run flawlessly with little intervention.
"An experimental design born by knowledgeable people is the fundamental building block of a stable process which leads to qualified and reliable heat treating processes"
Let’s straighten out the IOT timeline for heat treat processes. Twenty years ago, I visited a manufacturer of blades used in the logging apparatus. Their market had more than 1000 individual products. At the beginning of the day, operations would enter the part numbers of the previous day’s orders, and the rest was a connected manufacturing process producing a lot of 1 piece. I was genuinely impressed with this ultra-streamline method of responding to customer needs. Kind of sounds like the IOT dream before there was a name for it.
There was an assumption mentioned earlier that supercomputers deliver more reliable results faster. Well, I’m privileged to know some pretty impressive simulation modelling experts with some pretty remarkable computational hardware. Occasionally their intently reviewing their simulation results with both hands scratching the sides of their scalp. It passed through their mind that undetected solar irruption corrupted their files, or two electrons slipped through the processor unnoticed, heck they are moving so fast. Then calmer neurons prevail, and they review their data entry. Then you hear awe heck; it was in centimetres, not inches. To be more accurate in describing the power of the supercomputer, it should be stated when the data input is correct and the formulas verified in the previous study, then reliable results are produced quickly.
Back to my reason for being in that blade manufacturing plant a long time ago. Something changed in the heat-treating process on an intermittent basis. Two parts made hours apart from the same material had different hardness’. I was there to explain why this is happening and fix the issue. A metallurgist would probably conclude that with the same material and the same heat treating parameters, these two blades should have the same hardness. Walk a couple of doors down and ask the six sigma expert and ask ‘should two blades made from the same material have the same hardness’? Their reply will probably be in the form of two questions. Question one; was it the same part number blade? Question two; how many parts were in the sample to determine they were not the same hardness?
The issues of IOT, even with Industry 4.0 as work to standard for process control, cannot uncomplicate multidimensional cause and effect relationships as found in heat treatment. An experimental design born by knowledgeable people is the fundamental building block of a stable process which leads to qualified and reliable heat treating processes. Computers and Ethernet cables do not know their own. It’s time we spend more time talking about growing people as a most valued asset rather than connectivity rates and the internet.