There are good factories and then there is the best. The fine line demarcating these two is a universally accepted metric and renowned jargon in the corporate grapevine, known as Overall Equipment Efficiency (OEE). Often interchanged with the term Productivity, OEE, unlike the former has a narrower scope. OEE is closely hinged to a sub-process, equipment performance and the related factors whereas Productivity finds a rightful place among coalesce of several such sub-processes, like workshops or manufacturing establishments. Exceptional OEE of all equipment results in exceptional Productivity of the factory.
As one nears perfection, it becomes even more of a galling feat to achieve. Upgradation of equipment and profound industrial processes may make an uncanny difference in output but not at a penny-ante cost. There is always the last drop of efficiency that can be squeezed out before jumping onto the hard decisions. Identifying the right place to find this last drop is demystified with Industrial IOT (IIOT).
In numbers, OEE is the ratio of Fully Productive Time (FPT) to the Planned Production Time (PPT). Fully Productive Time is the total time required to manufacture faultless products. Several losses occur at different steps as we move from PPT towards FPT. These losses are broadly categorized into Availability Losses, Performance Losses, and Quality Losses. As we discuss these losses, we will find out how TOR IOT can help minimize them.
Availability Losses– Losses occurring due to unplanned downtimes are classified under this. These downtimes are a result of equipment breakdown or unplanned maintenance. The duration of the downtime is long enough for the operator to record the incident.
With IoT, availability losses can be easily avoided through Preventive Maintenance and Predictive Analysis. Web application compares the recent data of the machine with ‘control values’ using specialized algorithms and predicts possible failure modes. Deviations are highlighted and alerts are raised on frequent violations of the rules.
Subtle signs of misbehavior are missed due to inconsistent service schedules or plain negligence, leading to equipment failure. These failures can be avoided one step before Predictive Analysis through Preventive Maintenance. Maintenance scheduling ensures timely after-sale service of the machines and addresses the underlying problems in their infancy. Failures are prevented by generating alerts well before the conditions reach red.
Performance Losses– Frequent halts and sub-optimal utilization of the machine lead to Performance Losses. Halts are too short to record but frequent enough to affect the performance. Stops for refueling, lubrication, die setting and the like are indispensable to maintain quality but can be eliminated through the right procedures.
Frequent interruptions in operation is compared with the predefined limits, beyond which the notifications are pushed based on the priorities set.
Prime movers like electric motors and engines give the best efficiency at a fixed percentage of load and deviation towards either end of the spectrum disturbs the equilibrium but actual load during operation of the machine cannot be dictated with on-paper values.
Although efficiency in real-time cannot be measured with pinpoint accuracy, an acceptable operating range for a machine can be specified. Performance of every machine, when compared with the machines of similar type, provides relative effectiveness of each equipment. Performance comparison averages out any momentary slump in performance caused due to uncontrolled factors, which might affect one or several machines working in tandem.
Quality Losses– Quality losses include manufacturing defects and start-up failures. Every output failed reduces Fully Productive Time, thus affecting OEE. Manufacturing defects are usually process-driven and can be a result of incorrect settings, untrained operators, and other such factors.
These losses can be tracked at different checkpoints and diagnostics will help to identify the points with most failures. Once recognized, past records of the equipment at those points can be used to single out the exact cause.
In a nutshell…
Most industries adopt methods that involve human intervention. Maintaining manual records, periodic maintenances which miss out on minor glitches, overwriting or illegible handwriting in logs are just a few activities that are surprisingly prominent in well-established organizations to this day.
These actions may seem harmless at first glance with an argument that industries have sustained on these processes for decades.
Industrial IoT eliminates the human factor in monitoring. It gives crisp and clear information which is devoid of any ambiguity. By adopting IoT, the subtle differences get highlighted which were not possible to capture with conventional measures.
Furthermore, this data unlike previous processes do not stay buried in the archives. Analytics, resource management, production schedule can be derived from the recorded data which helps incorporate decisions being well-informed and backed with concrete numbers.
OEE has always been a sensitive subject and organizations have come a long way to earn their proverbial wisdom. What is good can always be made better and with IIOT, one can discover the ‘Hidden Factory’ within their plant with untapped potential waiting to be unleashed.
