Skip to main content

Changes in the mining industry

Published by
Global Mining Review,


‘Digital’ is a common word in the headlines of the global mining industry and, being such a broad term, Suresh Nagarowth, Vice President and Global Head of Delivery, Energy and Utilities, Cyient, would rather refrain from using that term by itself and would focus on the actual technology under that umbrella and the impact on the mining industry. Based on his conversations with leaders from various global mining majors, he has compiled three such changes and their drivers which are re-shaping the mining industry which, although not the three biggest changes in the mining industry, are definitely three important changes nevertheless.

Change 1: Mining majors are getting smarter and effective in the identification of new mining targets

Identification of new targets is of prime importance and mining majors want their geologists to have access to geophysical and geochemical data from pilot holes, legacy and public information at the click of a button. Traditionally, miners struggled in managing this data given the massive size of the data and a wide variety of data and storage formats. Now miners are leveraging technology to ensure that data is digitised, structured, indexed, easily accessible, and searchable. In addition, technology is used to extract insights to aid geologists in their job of establishing new targets. Few examples to illustrate the same include the following.

Use of AI/ML for classification of legacy geological records

Miners realise that legacy information stored across different servers and personal desktops of geologists has ‘nuggets of gold.’ Separating such information from a ton of other data including personal information is an exercise by itself. Recent regulations on privacy and GDPR add to the complication of using traditional manual and automation approaches for the classification of such data. From experience, data crawlers and search tools can help only to a certain extent. An effective approach for the segregation of useful information leverages machine learning techniques along with optical character recognition (OCR) technologies. A multi-modal cognition-based approach which includes language processing (word and flowing text) and visual cognition will help score the documents in terms of relevance and accordingly help segregate the information. This makes it cost-effective, faster, and efficient while ensuring privacy.

Digitised and searchable data along with identification of duplicates

Over the years, mining majors have accumulated massive amounts of legacy data that needs to be digitised and be made available to geologists. Miners are using a combination of smart OCR tools along with artificial intelligence/machine learning (AI/ML) based tools to extract data into a structured data model which can then pushed on to a central on-premise server or a cloud. The real value beyond the availability of structured data comes in identifying complete duplicates or near-duplicates as it allows to compress the data set and also ensures geologists do not waste their time on similar records.

UAVs and 3D mine models

UAVs provide easy access to remote areas and the ability to fit a wide range of imaging equipment makes them an attractive option for capturing right type of imagery. It starts with developing the required flight plan necessary for the acquisition of data specific to the problem. Then a host of custom and proven algorithms are deployed for data extraction, processing, integration, and finally quality assurance necessary. Subsequently, models including digital surface model (DSM) and Orthomosaics may be generated by using georeferenced data that is geometrically corrected and volumetric analysis of a dump. These 3D mine models can then help carry out volumetric analysis, disturbance analysis or environment management plan or management of waste dumps.

Change 2: Reduced operational costs through technology-led improvements in asset maintenance program

To extract the best value from an asset maintenance program, it is necessary to focus on the quality of maintenance master data and the maintenance strategy itself. Even though mining majors have for years used tools like SAP-PM or more recently tools like APM, quality of maintenance master data keeps haunting every operation or site because in simple terms ‘garbage in is garbage out’. Most of the issues are common across the industry and can be as elementary as null fields, incorrect hierarchy, and last but not duplicate part numbers. The challenge lies in the volume of data and lack of discipline while entering data into the tool. All of this leads to a wide range of downstream issues in the site which add up to a huge amount of unplanned downtime and high expenses in the form of poor spare parts management. This is where technology-led improvements are helping mining majors improve the quality of master data sitting in the enterprise asset management (EAM) system.

The second aspect of a maintenance strategy is equally important, and companies have evolved from traditional scheduled based maintenance to more recent condition-based maintenance or the latest predictive maintenance strategies. The essence of both condition-based and predictive maintenance strategy lies in the collection of data and timely decisions. If these two fails, then the strategy can lead to a situation where companies can lose more money compared to the traditional scheduled based maintenance. A few examples to illustrate how technology is helping mining majors in improving the quality of master data and effective predictive maintenance strategy are as follows.

Combination of latest technology-based tools to clean-up/migrate master data

An upcoming practice in the industry to improve the quality of master data is the deployment of rule engines. Such a rule engine(s) based tools crawl through the entire master data in the EAM system to identify potential errors and either fixes the same or propose a fix based on a set of pre-defined rules. It should be noted that the development of a rule engine of this kind, requires vast experience in master data management and is not a simple software solution as the strength of the solution is the robust set of rules for different conditions and not the software framework itself.

Another such custom framework or a solution set is being deployed for the effective migration of master data from one EAM system to another. Typically, it encompasses a five-stage process of extraction, consolidation, mapping, preparation of a loader sheet into a new EAM system, and finally validation of data

Such frameworks use a combination of traditional automation scripts, robotic process automation (RPA) techniques, and AI/ML based algorithms that leverage supervised learning techniques. Such solutions not only help cut down the time for migration of data from SAP to APM, but also ensure enhanced quality of master data once migrated to APM.

Implementation of next generation SCADA/DCS ‘integrated’ with IoT platforms for predictive maintenance of mining equipment

SCADA and DCS for mining fixed plants and engine management systems (EMS) for heavy mobile equipment can be helpful in reduction of maintenance costs. They provide valuable information to perform an RCA which then helps enhance the robustness of maintenance plans resulting in cost savings. They require plant and vehicle operators to closely monitor the alarms and a predefined set of reports to take an intelligent call to avoid potential failure. In addition to the lack of intelligence, the biggest disadvantage of these systems was limited ‘interoperability.’

There is a steep rise in the deployment of Internet of Things (IoT) platform-based solutions in the mining industry as these solutions make up for the limitations of SCADA/DCS and EMS. These platforms are driving both upstream and downstream integration of systems, are highly customisable and finally aid in decision making rather than solely depending on the call of the operator. However, the key to the success of these systems lies in the word ‘integrated’. Mining operators who have deployed these systems in pockets are struggling with limited benefits and the next big wave will definitely be towards the integration of the systems across the flow of inputs for predictive maintenance, maintenance planning, material planning, work order integration, and spare parts/ inventory purchases.

Change 3: Enhancing safety in mining operations by leveraging technology

Safety is of prime importance to mining operators and there is never a shortage of investment when it comes to enhancing the safety of personnel on the field. New generation trucks are now coming with advanced safety and driver-assist features like collision avoidance systems and active braking systems. The collision avoidance system (CAS) has particularly delivered some impressive results. For example, in its 2018 sustainability report, Newcrest reported that after implementing a GE Mining collision avoidance system, it saw a 33% reduction in vehicle-vehicle collision. Newcrest further went to state that they intended to deploy CAS in their other large mine site, Lihir. Beyond the equipment space, some more examples of how technology is being leveraged to enhance safety in mining operations are as follows.

Detection of hazards

In this case, there are two broad technologies, image detection and sensor-based detection. In terms of image detection, with a huge amount of surveillance camera data flowing into the central control rooms, mining majors are looking at leveraging image detection algorithms to identify potential hazards and raise necessary alarms to the operators. Algorithms can be used to detect the movement of people close to machinery or faulty operation of equipment that can lead to potential disasters. On the other hand, sensor-based systems have wide applications in predictive maintenance but in terms of prevention of hazards, they are used in the detection of loose parts which can fall of the equipment like buckets and chains. Timely detection of loss of contact is critical as it results in a massive accident beyond the significant loss of property.

Use of AR/VR/MR along with gaming techniques in operator training

Training is at the core of prevention of accidents and hence companies invest a lot in terms of workforce training. Augmented reality/virtual reality (AR/VR) have helped develop immersive training modules that help operators get an amazing orientation of the plant. Gaming techniques can be leveraged to make the entire training interesting and interactive. With the massive reduction in AR/VR set-up costs, these technologies have become affordable. Mixed reality (MR) makes it even more relevant as the operators can get a real feel of the actual site and operations. Just imagine a truck operator getting trained on the real truck cabin on the backdrop of a virtual mine site. It helps the operator get a feel for the site and the potential hazards during operation.

Use of wearables for detection and alarms

Wearables as a technology have been around for quite some time, but their usage has been relatively low. Technology advancements in the smart devices market have helped reduce the form factors of various chips and sensors that go on to wearables. Some of the common applications include gas detection and body vitals. There is a lot of development expected in this field and wearables would be a common aspect of field personnel.

As is evident, a lot of the technologies that come under the digital umbrella are being extensively used by mining majors’ in order to improve effectiveness and efficiency in mining operations. Adoption of these technologies is at rapid and mining majors are effectively partnering with technology and service organisations like Cyient to implement these technologies in their operations.

References

  1. Newcrest mining 2018 sustainability report.

Read the article online at: https://www.globalminingreview.com/special-reports/15102020/changes-in-the-mining-industry/

You might also like

 
 
 

Embed article link: (copy the HTML code below):