Throughout my career in Operations and Manufacturing I have essentially been tasked with getting information from the shop floor into the hands of managers and planners so that decisions could be made, and then getting the decisions back on to the shop floor so that they could be implemented.

It started fairly simple – PAPER.  We gathered information (e.g. production counts) off devices on to paper and sent them to an office where they were collated into one or more reports and passed onto management. Their decisions were passed back via production schedules, production meetings, or in some cases the board at the entrance to lets us know things had changed (we saw the changes as we walked into the plant).

In those days the dream was to get correct and accurate information and into the hands of those who need to perform their activities at “lightning speed”. That is, faster than it was possible.

Today the need to have the information in the correct hands is even greater as the velocity of business has increased tremendously.  With both old and new business processes depending on having information almost instantaneously, we are seeing greater focus on connecting “everything” to “everything else”,  that is the Internet of Things.

Connecting the devices, collecting the information and providing it to the business is speeding up operation capabilities in many areas.  For example being able to see current manufacturing activity (e.g. quantity produced, work-in-process, quality) can be used to accommodate  changes in demand that would have taken until the next reporting cycle (at the earliest end of shift), thus enabling better customer service.  In addition to what  I call the more traditional business processes (e.g. inventory management, production planning, quality management etc.) this flood of new information enables manufacturing / operations to consider new processes that where impossible due to the data  gathering and processing efforts.

One of our dreams when I was in operations was to be able to predict when a machine would fail, and under which circumstances would it fail faster.  We had a preventative maintenance program in place where we maintained the equipment based on run time, material produced, and the data collected about the equipment (e.g. visual inspection, some diagnostics). But nothing was really done about integrating this and other data to predict equipment failure.  Today we can (with the appropriate investment) eliminate the majority of the data gathering task for equipment with sensors can collect the data for us and pass this data onto other systems that can use this and other information gathered to answer questions that weren’t not even considered before:

Is there a correlation between variables, does one variable influence another? In one facility I worked it we were convinced (based on experience) that two products could not be run on parallel lines. But did we know why? If we could have found out the influencing cause via todays data collection and modeling capabilities we would have been able to increase the through put by resolving the problem. In another case we had one piece of equipment that never did operate as designed (we had not data collection capabilities in those days, everything was by personal experience), with an ability to gather data from and around the equipment we would have been more likely to resolve the problem.

Do we have a training problem? Are there a significant number of production problems or near misses related to scores in training, experience in operating equipment, hours on the job, time of year?  You probably of encountered the myth that cars built on the Monday and Friday had quality problems due to employees who were “still on the weekend” , but could the company prove that is was not true.

Can we improve cost of production by understanding current manufacturing costs? Traditionally the cost of manufacturing has been calculated at the end of production process, and sometimes as part of the end of the period process. Having the current operational information around manufacturing (e.g. material currently used, current operation parameters, scrap rates, labor hours booked, power being consumed, etc.)  could be of great benefit to production. It should be possible to calculate the profitability of making the product. Wouldn’t that be an interesting piece of information if you are asked to change the current production run. What might have been profitable could now generate a loss due to the requested change. And applying the current operating conditions to the planned production estimated profit can be calculated.  Thus we could go from managing costs to managing profit.

Would it be beneficial to have a customers’ demand changes to be automatically transmitted to you, and have your own networked equipment adjust to the new demand or specifications without your intervention? Farfetched, maybe, scary for some, but rapidly becoming technically possible.

Can you see the day where a piece of equipment that is having a problem, adjusts itself so that it can keep running, notifies maintenance that it is having a problem, puts in a request for a replacement part (based on self-diagnostics) and then notifies upstream and downstream processes that it is having a problem so that these other processes can adjust and the product still keeps being made. All without human intervention.

Does it help you to know where all your inventory is at all times? When looking at shipping or receiving inventory it is starting to become imperative that a company truly understands when a product is expected to arrive. Delays can impact production, customer service, and can generate large performance penalties.  Being able to see product shipping delays as they happen, understand changes in the shipping environment, identify unauthorized accesses and divergences all help ensure a quality product and service.  Without this connectivity we would still be doing investigation or adapting to changes after the fact and not as soon as the event happens

There are many benefits for connecting equipment & analyzing the data: early reporting of problems, predicting events, improving quality, understanding todays shop floor activities just to name a few.  We are entering a brave new world.  New processes are just being invented. We now have the ability to gather more data than we have ever imagined. We just have to decide what we are going to do with it.

To report this post you need to login first.

Be the first to leave a comment

You must be Logged on to comment or reply to a post.

Leave a Reply