Just knowing an alert has occurred and sending it to a human user may not be the ideal outcome. Again, just like in the previous sections, depending on humans to handle tasks introduces variability and more important, excess cost. Let’s look at an example of a machine that is experiencing abnormally high hydraulic temperature. We have determined this is abnormal based on the specific machine, the in context item / part specific parameters and perhaps external sensors such as ambient temperature. An alert gets raised and is sent to the maintenance team. At this point in a manual process, the maintenance team will create a workorder for service and schedule a person to service the machine’s oil cooler. Frankly we can do better than this. Why not have the alert generate a workflow that automatically creates the workorder and based on a rate of temperature rise on the machine (since we have that data as well) places a priority as to how quickly to service the machine. Perhaps the temperature has been slowly rising over the past 2 weeks – less of an issue than if it skyrocketed in the last hour and we have an immediate need to avoid crisis. Workflows should be used to streamline tasks of people and reduce their human effort to manually process activities. Industrial oriented IoT solutions will need to have industrial based workflow solutions that can make decisions based on the IoT data that has been collected.
Bottom Line – just having alerts is not enough to satisfy the needs of a proper IoT enabled manufacturing solution. Having integrated workflow capabilities that can consume IoT in context data, apply rules and process paths based on the data and then drive downstream systems such as maintenance, quality, front office ERP activities will bring much higher value and use.
Stay tuned for part 5, "Analysis and Reporting"
Learn more at www.Godlan.com