Predictive maintenance analytics to reduce overall downtime for a leading fertilizer company

Company Overview

The client is a leading fertilizer company and environmental solutions provider in the US. As part of their production process, they deal with a large number of machineries, including rotating equipments.



Problem Statement & Challenges

The client didn’t have a contingency plan if a component or item needed replacement, leading to failures in their rotating equipments.

As a result, the client observed several unplanned outages.



Pain Points
  • Unplanned shutdown
  • Unable to forecast an equipment/part failure


Objective

To build a sophisticated predictive maintenance model which includes the calculation of Remaining Useful Life (RUL) of the equipment.

With the help of RUL -

  • Engineers can schedule maintenance
  • Optimize operating efficiency
  • Avoid unplanned shutdown


Our Solution/Approach

Remaining Useful Life of an equipment was estimated from proportional hazard models and probability distributions of equipment failure times. This was done by determining the failure time of a component based on past failure times and covariates, variables, for instance, the environment in which the equipment operated (such as temperature, pressure, etc.) were also considered. The model could predict the probability of a machine if it is at the end of its lifetime based on its operation cycles.

The estimates for the Remaining Useful Life were integrated into dashboards and existing alarm systems that were monitored by the maintenance team. Using these dashboards, the team were able to respond to changes in equipment health as quickly as possible.



Results / Impact

The client was able to achieve the following results –

  • Engineers could conduct scheduled maintenance; plan for spare parts in advance
  • Optimize operating efficiency
  • Reduce overall downtime
Casestudy Infograph Fertilizer