Energy-Centric Predictive Maintenance Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current preventive maintenance scheduling for assets does not effectively consider specific site conditions or energy wastage, often resulting in unnecessary costs and downtime, as it is based on routine intervals rather than asset performance and energy efficiency.
Innovation Solution
A method using a data model, such as a Temporal Fusion Transformer deep learning model, to predict sensor state values and energy usage, calculating energy wastage, and recommending maintenance tasks based on when the wastage equals the cost of the tasks, thereby optimizing maintenance scheduling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If preventive maintenance is scheduled based on routine intervals, then asset availability is maintained, but maintenance costs increase and energy wastage occurs due to unnecessary maintenance tasks
Solution Approach 1:
The maintenance schedule transitions from static routine intervals to dynamic scheduling based on real-time asset performance data and predicted energy wastage, allowing the system to adapt maintenance timing to actual asset conditions and energy cost variations
Solution Approach 2:
The system changes the parameter basis for maintenance scheduling from fixed time intervals to variable parameters including asset performance metrics, predicted energy wastage, and maintenance cost thresholds, enabling optimization of both reliability and energy efficiency
2Reliability
If preventive maintenance is performed more frequently, then asset reliability is improved, but maintenance costs and downtime increase
Solution Approach 1:
The system performs preliminary analysis of asset performance trends and predicts future energy wastage and maintenance needs, allowing proactive scheduling of maintenance tasks at optimal times before actual degradation occurs, minimizing unnecessary downtime
Solution Approach 2:
The system continuously monitors asset performance data and uses feedback loops to adjust maintenance scheduling decisions, comparing actual asset conditions against predicted thresholds to determine the optimal timing for maintenance tasks
3Loss of energy
If maintenance is delayed, then maintenance costs are reduced, but asset deterioration increases leading to unplanned downtime
Solution Approach 1:
The system performs preliminary prediction of asset degradation trends and energy wastage accumulation, identifying the optimal point to schedule maintenance before asset deterioration reaches critical levels, balancing cost reduction with reliability maintenance
Solution Approach 2:
The system replaces traditional mechanical time-based maintenance schedules with an intelligent decision-making system that uses data analytics, machine learning models, and energy cost calculations to determine optimal maintenance timing
Data Source
AI summary
Methods, apparatuses, and computer program products for energy-centric predictive maintenance scheduling are provided. For example, a computer-implemented method may include inputting historical time-varying sensor state values associated with an asset into a data model to train the data model; inputting expected future time-varying asset-independent data over a time frame into the data model; generating from the data model predicted sensor state values and energy usage associated with the asset over the time frame; determining optimum energy usage by the asset over the time frame; calculating energy wastage over the time frame based on a difference between the predicted energy usage and the optimum energy usage; calculating, using the predicted sensor state values, one or more asset performance metrics corresponding to one or more preventive maintenance tasks; and generating and reporting one or more recommended service tasks over the time frame based at least in part on the calculated energy wastage.