Skip to main content

Predictive Maintenance Driven Task Planning System

This case-study proposes an optimization model that integrates predictive, scheduled, and daily maintenance tasks in manufacturing to assign technicians efficiently and minimize costs while maximizing task completion.


Business Category
Technical Category
Planning and scheduling


Predictive maintenance and energy management systems play a crucial role in forecasting maintenance needs in manufacturing environments. While scheduled routine maintenance and everyday operational tasks are standard in these settings, there is a distinct absence of an in-depth study that combines insights from these systems. Such a study could significantly optimize daily task scheduling for planners, offering a more efficient and effective approach to managing manufacturing processes.

Problem Statement

The problem involves the efficient assignment of qualified technicians to machines requiring different maintenance types. This challenge encompasses three types of tasks originating from maintenance operations. The first type involves scheduled tasks, which include regular maintenance and inspections. Additionally, there are AI-driven predictions for remaining asset lifetimes, constituting the dynamic second-type tasks. The third type involves urgent tasks that emerge from breakdowns and are identified through anomaly-detection models. Given a set of machines, maintenance types, technician qualifications, technician working hour shifts, and skip cost data, the objective is to determine the optimal technician assignments and identify any skipped maintenance, while minimizing the overall cost and maximizing the number of tasks assigned. To adress this issue we have developed an optimization model that is fully integrated with the predictive maintenance, scheduled maintenance and daily maintenance task of the system.


In our case study, we employed several Key Performance Indicators (KPIs) to gauge the effectiveness of our model:

  1. Machine Downtime:

    • KPI: Time machines are inactive due to maintenance or breakdowns.
    • Goal: Increase equipment uptime by 10-15%.
    • Progress: Achieved a 66% reduction in breakdowns, significantly increasing uptime.
  2. Cost of Spare Parts and Consumables:

    • KPI: Expenditure on spare parts and consumables for curing presses.
    • Goal: Reduce costs by 10-15%.
    • Progress: Attained an 80% reduction in spare parts costs.
  3. Maintenance Planning Time:

    • KPI: Time spent on maintenance planning.
    • Goal: Cut down planning time by 60-70%.
    • Progress: Reduced planning time from 3-4 hours to 10-15 minutes.
  4. Breakdown Rate:

    • KPI: Frequency of breakdowns during production.
    • Goal: Decrease breakdown rate by 5-10%.
    • Progress: Lowered breakdown rates by 5.1% in the selected period.