Operations and Maintenance managers are under constant pressure to make sure that plants are running at full capacity. Unforeseen failures and unplanned downtime are a challenge. Their responsibility is to plan appropriately and to choose the right maintenance strategy and planning. This first blog on Predictive maintenance (PdM) will discuss common maintenance challenges and how they can be mitigated.
Asset-intensive industries face significant challenges during the maintenance of assets, impeding optimal performance, increasing costs, and affecting operational efficiency. These challenges encompass aging infrastructure, high maintenance costs, complex asset structures, regulatory compliance requirements, resource allocation difficulties, safety and risk management concerns, asset performance optimization, and the skills and knowledge gap in the workforce.
Resolving these challenges is crucial to ensure uninterrupted operations, minimal downtime, and maximum asset utilization, but it requires comprehensive solutions that address each specific challenge and integrate them into a cohesive maintenance framework tailored to the asset-intensive needs of the critical assets.
Maintenance can be defined as set of activities and processes to make sure that assets function optimally thereby attaining a desired level of productivity. All physical assets degrade over time due to wear and tear of parts and therefore require maintenance. Maintenance strategies should align with the long-term goal of the enterprise because it is expensive, time-consuming, and may result into unplanned interruptions if not done properly.
Asset uptime is at the topmost priority of asset-intensive industries such as oil & gas, metals & mining, power, chemicals, petrochemicals etc. Asset downtime can cause huge productivity losses whilst maintenance of assets could be as high as 20% to 60% of the opex spend depending on industry, asset type and capex spend.
According to an article published in March 2021 by the news magazine Manufacturing Global, 98% of businesses report that a single hour of interruption affects their productivity, costing them more than US $100k.
The cost of downtime can be as high as US $1.8 M/ day in a 200K barrels/day refinery capacity. In oil drilling the cost of non-productive time per asset from drilling to completion is US $500K to US $1M per day; post-completion is US $40K to US $300K per day on average. According to data collected in a survey by Vanson , 82% of companies surveyed across industries were hit the hardest financially by unplanned downtime. These are only a few examples to illustrate that without a proper maintenance strategy implementation, organizations will not only lose money but could lose goodwill and reputation if there are safety risks and environmental damages.
Type of maintenance strategies
Maintenance strategies can be broadly classified into a Reactive and a Proactive approach. The table below provides an overview of the spectrum of maintenance approaches from Reactive maintenance on one side to Prescriptive maintenance on the other side of the spectrum. The more mature a Proactive maintenance approach is the more it relies on digital technology such as sensor technology, IOT platforms, AI and machine learning, and the availability of quality historic and real-time data.
Why PdM? Over the last two decades, there has been huge advancement in the computational capabilities, IoT, connectivity, data storage and processing, cloud, and artificial intelligence. Together with a tremendous drop in computational -and IT infrastructure costs this has unlocked the capability to predict equipment failure, reduce unplanned downtime, reduce maintenance cost and increase asset life. With the latest trends in mind, adopting a predictive maintenance approach may be the right choice for asset intensive industries.
Many organizations in these asset intensive industries have started to leverage the benefits from PdM. Predictive maintenance includes use of wireless data collection, data analysis, pattern recognition and making predictions about an asset’s health. Predictive maintenance techniques use the machine learning models to find the anomaly using historical data as well the real time data from an asset. The data used for modelling could be operational (such as thermal, oil analysis, infrared, vibrational or acoustic etc.), maintenance logs.
Below are some of the benefits that can be realised by implementing predictive maintenance solutions. There are several factors that could affect the overall success but if implemented properly, organizations can benefit by reducing costs and improving efficiencies.
This blog covered an introduction to maintenance and why it is important for organizations to implement proper maintenance strategies. Many organizations have started implementing data driven maintenance strategies such as predictive maintenance. In our second blog we will discuss the role of digital and data in predictive maintenance implementation.
Oliver Van Belle