Big data and the waste industry

Columns - Waste Watch

Subscribe
August 2, 2019

The buzz around the internet of things (IoT) and big data is a bit like a shotgun blast: loud, far-reaching, but all over the place. Despite the fact that these terms have become a common part of the lexicon, it is difficult for people in certain industries to understand how these technologies affect them. This is true in the waste and recycling processing sectors, where applying cutting-edge technology is often an afterthought, even in the best of times. Regardless of one’s stance regarding this new wave of information, there are opportunities for industry participants to integrate these solutions to their advantage.

One of the most likely areas where the solid waste industry may use big data is in the transition from preventive to predictive maintenance for fleet vehicles and processing equipment. Preventive maintenance refers to the general upkeep of equipment before system failures occur. This could include tasks such as greasing the bearings of a conveyor at regular intervals or changing the air filter of a collection truck after a certain number of miles, regardless of whether the filter actually shows signs of wear. Predictive maintenance, on the other hand, requires past data to forecast the estimated time of failure for a part or equipment. This could include replacing a conveyor belt after a certain amount of material is transferred or using testing to predict when vehicle oil should be changed based on the number of engine hours and usage duty. While both the use of preventive and predictive maintenance can reduce catastrophic equipment failures, when used together, the two methodologies can transform maintenance programs to be more needs-based and efficient. A maintenance program that uses both preventive and predictive methods can save time, money and downtime by resulting in maintenance only when called for based on environmental factors and real-world usage patterns.

Predictive maintenance can be a very useful tool in the waste industry for reducing costs in fleet maintenance, rolling stock and processing equipment, but this type of information can also be applied to carts and bins to begin to predict fill levels and replacement or refurbishment schedules. While the increase in efficiency is one of the largest payoffs associated with predictive maintenance—as are the reduction of equipment failures and downtime—it is dependent on the collection and interpretation of many data points, which is why this methodology is still something rarely applied in the waste industry today.

Currently, the application of IoT technology has several shortcomings, including in the waste and recycling industries. First, much of the equipment currently used in the industry does not record the data necessary for the proper interpretation of predictive maintenance. Second, even if one does have these data points available, there are few software application options on the market that are able to collect and interpret the data in a manner that allows the user to identify the patterns necessary for predicting equipment and part failure.

With these glaring shortcomings, the question remains: What can solid waste and recycling managers do in the interim to be ready for the inevitable future of big data? Currently, all entities in the waste and recycling industry should anticipate that the ability to use IoT data to create custom predictive maintenance procedures will be available in the near future. Therefore, the equipment purchased now should be data-ready. This means new equipment should be specified with data-producing hardware, such as on-board computers, radio-frequency identification (RFID) tags or motion sensors to record usage. This will make the transition to system readers and data interpretation much easier once this technology comes online. No one wants to build a processing plant from last century, so it’s important that procurement specifications ask for the equipment that can best see our industry into the future, not keep us where we’ve been.