As we approach the end of the year we’re taking a look at some of 2020’s plant failures within Heavy Industry, and ranking each one on its news, reputation and cost impact. As well as how preventable it may have been. Which ranking will you choose to compare?
Madukkarai Cement fined US$61,000 for fugitive cement dust1
In mid-2020, ACC subsidiary Madukkarai Cement was found to be releasing cement and clinker dust from its plant. After complaints and protests from residents nearby, an ambient air quality survey in September revealed higher-then-prescribed particulate volumes. The company was fined $61,000 and must meet 19 standards relating to air pollution control and monitoring by early January 2021.
The manufacturing sector is subject to increasingly strict environmental standards and harsh penalties can be imposed for failing to meet them. Meeting air pollution targets with confidence requires accurate measurement, typically using optical and particle sensors. Maintenance schedules designed to prevent breakdowns aim to address common issues without any consideration given to actual equipment health and stresses. In contrast, predictive maintenance solutions that use machine learning and real-time sensor data, historical data and manufacturer information to assess machinery health and identify unusual behavior can predict failures before they happen.
Explosion at US paper mill2
In April, a paper mill in Maine, US suffered an explosion, destroying a large part of the mill and stopping pulp manufacturing for months, leading to workers losing their jobs. The explosion was caused by a ruptured pressure vessel in the mill’s pulp digester, which mixes wood chips, water and chemicals into pulp that eventually becomes paper.
This explosion had a huge impact on the running of the paper mill, with long-lasting damage and staff layoffs. Fitted with suitable sensors, such as vibration monitors, the pressure vessel is likely to have shown signs of wear which a predictive maintenance approach could have identified in real time for further investigation. Incidents such as sudden ruptures are very difficult for a preventative, interval-based maintenance schedule to manage.
Central Asia Metals may lose zinc, lead output for a month on tailings dam leak3
In September, Central Asia Metals’ North Macedonia mine experienced a tailings dam leak, leading to a shutdown of the processing plant and its mining activities lasting several weeks. The leak lasted for approximately 1.5 hours before being halted. Tailings dam safety has become a global topic and the company was required to undergo an investigation into how the leak occurred, with dividend payments being withheld as a result. With the negative ESG press impacting investment, a possible financial penalty and operational and compliance changes, the repercussions of this incident could continue for some time.
Tailings dam failures can be caused by overloads, anomalous behavior of the material used to build the dam (normally tailings), or from problems with the drainage systems resulting in an increase of pore water pressure, and therefore a loss of resistance. Understanding critical failure modes is key for long-term risk planning. Real-time data gathering enhanced with advanced analysis techniques means that a predictive maintenance solution can model the dam’s normal behavior and quickly flag up any unusual data before the dam fails.
The winners and losers
So, which ranking did you choose to play with? The headlines may spark varying levels of interest, the reputation scores high and low, but the preventability scores across the board are all high. Moving to Industry 4.0 is easier and cheaper than ever, with low-cost sensors available to retrofit to existing machinery and cost-effective, cloud-based PdM solutions for fast, scalable results. Mass failures don’t need to happen.
Senseye ROI Lock®
Senseye now independently guarantees a reduction in machine downtime for industrial and manufacturing companies. ROI Lock is supported by a Tier One reinsurance partner with AA- rating, guaranteeing that your predictive maintenance rollout has a predictable, 100% minimum Return on Investment.
References: