Predictive Maintenance in Forging Industry

Abstract

Overall Equipment Effectiveness (OEE), which is a standard to measure equipment productivity, heavily depends on the efficiency with which it is working. Any downtime of equipment will immensely affect the efficiency and subsequently affect the OEE of the equipment. In a production line, this unplanned downtime will have repercussions over other work centres causing a catastrophic effect on the factory's throughput. If such downtime occurs on any of the bottlenecks, the entire factory remains shut, with throughput becoming zero until the issue at the jam is resolved. To avoid such equipment failures, it is imperative to conduct maintenance of the machines in the facility at subsequent intervals. Although timely maintenance has a cost attached to itself, it certainly does save a lot more than the downtime cost incurred during a failure. With that being said, the cost is not the only concern concerning a machine failure; improper maintenance can even have safety concerns and can cause injuries to workers, subsequently leading to legal issues. Hence, timely equipment maintenance is of paramount importance for any state-of-the-art facility to remain a global leader in the market. Predictive maintenance is a method that can be done by utilizing the various analytics and machine learning tools that help predict with accuracy when a machine requires support. A viable PM model, with the utilization of memorable information of disappointments, assists us with foreseeing the time at which a machine will run into disappointment.