A spatio-temporal approach to underground mine health and safety monitoring

July 30th, 2019, Published in Articles: PositionIT, Featured: PositionIT

Health and safety of employees in their work environment, particularly in underground mines which are deemed one of the most dangerous work places in the world, is of critical importance. This article highlights a prototype application that integrates legacy underground mine systems with a geographic information system (GIS) for the analysis of streaming digital sensor data.

The mining industry has been a major driving force behind the development and advancement of South African economy. The City of Johannesburg, for instance, resulted from the discovery and mining of gold in the early 1880s. Johannesburg has grown from a informal tent/hut settlement to a formal city within five years and to its present state as a metropolitan city [1]. South Africa is mineral rich, accounting for approximately 11% of the world’s gold and 40% of the world’s known resources, and the industry directly employs about 500 000 miners with about 4,5-million dependants [2]. These figures represent a labour-intensive industry, which reiterates the importance of a healthy and safe environment for underground miners [3]. Underground mine incidents, accidents and health risks have significant consequences, including the closure of mines, financial loses and even death of mine personnel. Therefore, an initiative to monitor and manage these health and safety risks and factors is called for, especially with the advancements made so far in technology.

There have been tremendous developments in information, communications and technology (ICT). Examples include faster data processing speeds, increased storage capacities by computers and mobile devices, information extraction capabilities through machine and data learning, ultra-fast communication technologies through wired and wireless technologies, and development of micro-sensors for data acquisition, to mention a few. The adoption of technology has also increased, enabling not only individuals but also the development and integration of more robust and efficient systems. Such systems have found their way into underground mining, which has long neglected or underutilised ICT systems and implementations compared to quarries, surface and open pit mines. These include heating, ventilation and air conditioning (HVAC) systems for ventilation and refrigeration, proximity detection with collision avoidance systems, and Internet of Things (IoT) sensors for real-time data collection and communication, among others.

Accidents and incidents in underground mines are in most cases the result of natural occurrences, human errors or mistakes and machinery malfunctions. Depending on the mineral ores being mined, the depths of a mine facility and its mechanisation levels, miners are faced with four prominent risks and hazards: underground fires, mine collapse or rock falls, drowning and exposure to toxic and dangerous gases and contaminants. The industry has experienced disasters in underground mines, with some of the recent reported incidences in South Africa including a fire at Kusasalethu Gold Mine on 24 February 2014 from sparks while undertaking a maintenance exercise; a shaft subsiding at Sibanye Stillwater’s Driefontein mine on 3 May 2018; and exposure to heat and toxic gases at an abandoned shaft at Sibanye Stillwater’s Kloof Ikamva mine.

The underground mining industry has made progress in ensuring the facilities and work environments are healthy and safe for miners, but due to the nature of some of these incidents and difficulty in accurate prediction of such fatal incidents, the need for a continuous monitoring system is needed to provide a means for early detection, warning and minimisation of risks and fatalities that would result from such an impending disaster.

The adoption of technology and smart systems in underground mining however has also brought with it a problem of silo systems. Different vendors are developing and deploying proprietary systems, with some of these systems not having the capabilities for data and information sharing with other systems. Developing systems capable of integrating with others allows for efficient acquisition and implementation of systems, analysis of collected data from all the systems and derivation of in-depth information about a facility and environmental situation in an underground mine.

Another problem involves delays in data processing and relaying information. The difference between life and death or making timely, intelligent decisions when faced with health and safety hazards can be as little as a few seconds. Real-time data analysis and spatial reporting in an underground mine would be beneficial in making location-based, intelligent and timely decisions.

Real-time monitoring and analysis with GIS

Real-time systems are defined as systems that are capable of processing its input data and report back its results within a defined, often short, period of time from receipt of the input data [4]. Continuous and real-time monitoring and measurements of environmental factors, locating of static and mobile machinery (including personnel tracking in underground mines), are necessities in the modern mining industry.

A geographic information system (GIS), on the other hand, is used for the collection, storage, updating, analysis and visualisation of spatial data. Spatial data is an integration of location and associated attribute information about a geographical feature or object on, above or below the earth’s surface. The ability of a GIS to manage and analyse spatial data is the main reason why it is well suited for systems integration and visualisation of “space”.

A system for health and safety monitoring within underground mines can be achieved by integrating and analysing real-time data on a GIS platform. This article highlights a prototype application that integrates legacy underground mine systems with a GIS for the analysis of streaming digital sensor data. The system is developed on Esri’s ArcGIS platform with the sensor data collected at the Wits Mining Institute’s DigiMine mock underground mine facility, which is also equipped with a control room and digital sensors system collecting environmental parameters such as temperature, humidity and dangerous gases. The WMI mock underground mine also has sensors for monitoring and recording seismic events and rock movements, including the recent earthquake of 2018 with its epicentre in Botswana, which provides valuable data for research at the DigiMine facility.

Fig. 1: WMI Mock Underground Mine, located at the Chamber of Mines building at Wits University.

The prototype system implements a client-server architecture, achieved with the implementation of Microsoft SQL Server Express 2012 as a relational database management system (RDBMS), and ArcGIS for Server with ArcGIS GeoEvent Server extension for real-time processing of the streaming sensor data [5]. An operations dashboard application was also developed to provide a visual and graphical interface to interact with the integrated system.

Fig. 2: The tunnel section of the mock underground mine facility.

Integration with digital sensors for access to streaming sensor data and their processing was achieved through the development of a GeoEvent processing service. This service is a configuration of data input, data processing and data reporting services, done through a GeoEvent Manager application. The GeoEvent service’s input and output configurations involved developing a data schema for the streaming sensor data with links to the database for storing the processed data and information. The service also contains data filters for the streaming sensor data input, geometry and conversion of the data into spatial information, and a field mapping functionality to enable linking the feature service layers and the database fields. A snapshot of the GeoEvent processing service showing the process workflow is illustrated in Fig. 3.

Fig. 3: A GeoEvent Service workflow for sensor data processing.

Data reporting and visualisation

A dashboard application was developed for the visualisation of the filtered and processed streaming sensor data. This interface provides for a graphical and GIS-based visualisation platform with colour-coded gauges and alerting tools to communicate information about the underground facility.

This operations dashboard interface is both mobile and desktop-ready and can be accessed and used within an enterprise. The results from data learning and derivation of information from the data stored in the database is also made available and can be visualised in the dashboard, as shown in Fig. 4.

Fig. 4: An operations dashboard application.

Conclusion

Underground miners are faced with serious dangers, hazards and health risks why plying their trade and profession kilometres below the surface. Through integrated systems, mine monitoring systems can be developed and implemented alongside digital sensors for round-the-clock monitoring and management of a facility and its environment.

Such an integrated system, developed on a GIS platform and with an operations dashboard included as an interactive and graphical interface, provides additional benefits such location awareness to mine monitoring. This enables making localised decisions affecting specific regions within the underground mine, such as determining danger zones, exit routes, the location of the nearest emergency equipment, location and personnel count within certain sections of the mine, just to mention a few, thus enabling efficient emergency response and rescue systems.

Machine learning or data learning and information extraction system bring further capabilities for risks prediction and early warning [6, 7]. These can be derived by analysing historical data and determining trends in the data. This information can be used to determine and classify, according to danger levels, the prevailing and predicted future conditions of an underground mine environment.

The inclusion of real-time processing and communications are also essential in providing timely feedback and warnings to miners deep in the mine. This prompt alerting or warning system will contribute to the reduction (and even elimination) of casualties in underground mines.

Health and safety of mines are of critical importance and technology provides a means by which lives can be preserved while ensuring working conditions in underground mines are favourable.

References

[1] K Nhlengetwa and KAA Hein, 2015. Zama-Zama mining in the Durban Deep/Roodepoort area of Johannesburg, South Africa: An invasive or alternate livelihood? The Extractive Industry and Society, 2(1), pp. 1-3.
[2] Chamber of Mines, 2017. Chamber of Mines – Quarterly Update for February 2017. Available online: http://www.chamberofmines.org.za/industry-news/publications/newsletters/send/13-newsletters/404-quarterly-update-february-2017 [Accessed February 23, 2017]
[3] B Hebblewhite, 2009. Mine safety-through appropriate combination of technology and management practice. Procedia Earth and Planetary Science, 1(1), pp. 13-19.
[4] SM Peter, 2008. Real Time Systems. Available online: http://www.cse.unsw.edu.au/~cs9242/08/lectures/09-realtimex2.pdf
[5] R Lamar, 2014. Improving Health and Safety with Access to Real-Time Events using GeoEvent Processor. The Pennsylvania State University. Available at: https://gis.e-education.psu.edu/sites/default/files/capstone/Lamar_596B_20141215.pdf
[6] B Jo and R Khan, 2018. An Internet of Things System for Underground Mine Air Quality Pollutant Prediction Based on Azure Machine Learning. Sensors, 18(4), pp. 930.
[7] S Verma and S Chaudhari, 2017. Safety of Workers in Indian Mines: Study, Analysis and Prediction. Safety and Health at Work, 8(3), pp. 267-275.

Contact Calvin Oduor Opiti, Tel 081 062-2237, opiticalvin@gmail.com

Related Articles

  • Hackathon prepares learners for fourth industrial revolution economy
  • Geospatial information is crucial for Africa’s economic development
  • South African engineering excellence celebrated
  • National development plan to be reviewed
  • How to create modern data systems for sustainable development