IT Trends in the Mining Industry
“Today, miners are operating highly technical and complex machines that are worth many millions of dollars.”
In the past decade, China has been the world’s largest consumer of metals and minerals and major growth in the Chinese economy fueled a “mining super cycle.” Now the reduction in the growth rate of the Chinese economy has triggered the current “super cycle” of contraction in the mining industry.
Apart from cyclicality, there are two other dominant characteristics of the mining industry, namely capital and safety. Modern mining is no longer a people intensive industry. Today,miners are operating highly technical and complex machines that are worth many millions of dollars. Mining is a very capital intensive and safety-conscious industry. Think of mining along the same lines of aviation and nuclear power generation…they all have a similar focus on safety.
In the up cycle, the primary focus of every mining company is output in order to meet the demand from their customers. So operator productivity, machine availability and performance are the key business metrics. Then in the down cycle, the drive for maximum output is replaced by a drive for maximum productivity in order to lower the costs of production. This means that operator productivity and machine productivity become the key business metrics.
Clearly, the ability of the IT function to acquire operational data and then process it into management information in a timely manner is critical to the success of any modern mining business. Data flows from the machines to the IT side of the business and, ultimately, to the customers.
Modern mining machines are operated via digital control systems that can generate data, and the high capital values of mining machines (shovels, trucks, loaders, longwalls etc.) means that the incremental cost of “IP enablement” is not an issue.
Thus the first IT issue to decide is what data should be captured from the machines. What should the sampling period be (a voltage spike could last less than one second but temperature would take several minutes to change noticeably) and how frequently should the data be transmitted? Should regular time based data be captured, or only exception-based data when some parameter goes outside the predefined “normal operation limits”? Operator performance data usually has a longer sampling period and lower transmission frequency than machine-based data.
Efficient data compression (usually via a data logger) is required in order to minimize the data that must be transmitted. And transmission is usually a challenge for the IT function. Mines are a “hostile environment” for IT equipment, especially underground mines, and most are typically located in places that have little or no network connectivity available. And the network will probably have to manage both continuous data streaming and regular large file transfers.
Assuming that the network challenges have been met, the IT function will then need to handle very large volumes of data. Control systems could have as many as 4,000 data tags on the most complex equipment. And the data needs to be rapidly analyzed, with information being fed back to the mining operations teams as fast as possible.
Information on operator performance, for example, can be used to identify areas where further training is needed to reduce errors, increasing operator safety and efficiency.
The main objectives of machine-based data are early fault detection and accurate preventative maintenance. It is possible to use the data and IT tools to predict when a part should be replaced in order to prevent a major machine failure that would significantly increase the cost and time needed to repair.
Because Joy Global is a direct service company, engaging directly with mining customers worldwide, we have a viewpoint that spans multiple mines for multiple customers in several regions. This allows us to generate benchmarking data. For example, we can calculate the average and “best in class” productivity of a copper mining shovel in Latin America and use this information to help improve the performance of other shovels, enabling“best in class” operational performance.
We can also analyze the performance of individual machines in order to optimize an overall process. For example, wait times for a shovel could indicate that the ratio of trucks (move) to shovels (dig) is not optimized and that more trucks, a shorter drive distance or a conveyor belt are required to enhance productivity.
Ultimately, the mining industry is driving toward full automation, which will remove people from harm’s way. The miner will operate the underground mining equipment from the safety and comfort of an “above ground” office complex, which could be at the mine or somewhere else. In order for this to become a reality, the IT function will need to deliver totally reliable real time networking capabilities, or a level of machine intelligence that will allow the machines to operate for a short period of time without input from a human operator.
Mining will probably always be about people working in tough and inhospitable conditions. But IT has, and will continue to, transform the industry. Mining CIOs will need to be closely involved with product engineering to create data, but also with internal and external customers who consume IT services based on this data.