top of page

Mine Guard AI at QMIHSC24

Writer's picture: ekin eraydinekin eraydin

Updated: Sep 15, 2024


At this year’s Queensland Mining Industry Safety and Health Conference, we had the exciting opportunity to present the potential of Generative AI in transforming the way incident investigations are conducted. In a sector where safety is paramount, leveraging cutting-edge technology like AI to drive efficiency and accuracy can have a profound impact. Our presentation at the conference focused on the strengths and weaknesses of using AI in safety investigations, illustrated through a live case study, and ended with a discussion of future opportunities for this technology.


Strengths of Using Generative AI in Incident Investigations

We kicked off the presentation by highlighting the core strengths of using Generative AI for incident investigations. One of the key advantages is the ability to automate complex and time-consuming tasks, such as generating accurate event descriptions, identifying contributing factors, and providing follow-up interview questions. This automation not only reduces investigation time but also enhances the overall quality of the analysis.

Moreover, AI can significantly minimize human bias during investigations. Investigators, regardless of experience, are prone to certain biases that can affect outcomes. By relying on AI to process and analyze data impartially, ICAM Expert delivers more objective insights, ensuring investigations are consistent and standardized across the board.


Full Presentation:



Weaknesses and Areas for Improvement

Of course, no technology is without its challenges. We openly discussed the potential weaknesses of AI-driven incident investigations. One notable limitation is the current inability of AI to fully grasp the contextual nuances that human investigators bring to the table. While AI excels at analyzing large volumes of data and identifying patterns, it still requires human oversight to make sense of more complex situations that involve subjective judgment or unique operational circumstances.

Additionally, while AI reduces human error, it is dependent on the quality of the data fed into it. Poor data quality or incomplete information can lead to inaccurate results. To mitigate this, we emphasized the importance of thorough data collection practices to ensure that AI has the best possible input for processing.


The centerpiece of our presentation was a live case study showcasing our AI-powered tool, ICAM Expert, in action. We demonstrated how the tool can take real-time data from an incident, process it, and deliver a comprehensive ICAM (Incident Cause Analysis Method) report with minimal human input. This live demo allowed the audience to see firsthand how AI can handle tasks like root cause analysis, PEEPO (People, Environment, Equipment, Procedures, and Organization) analysis, and contributing factor identification—all with speed and precision.

The case study emphasized how ICAM Expert can reduce investigation times, helping mining operations quickly respond to incidents and take corrective actions. The audience was particularly interested in how AI can be used to standardize investigations, ensuring every incident is approached with the same thoroughness, regardless of who is leading the investigation.


Future Opportunities

We closed the session by discussing the future opportunities for AI in incident investigations. As the technology continues to evolve, we foresee AI playing an even greater role in predictive safety management. By analyzing historical data from past incidents, AI could help identify potential risks before they materialize, allowing companies to take proactive measures to prevent incidents altogether.

Another exciting area for growth is the integration of AI with other safety systems. For example, combining AI with real-time monitoring tools could create a comprehensive safety ecosystem that not only investigates incidents but actively monitors operations for signs of risk. This could further streamline the journey toward becoming a High Reliability Organization (HRO), where the focus shifts from reacting to incidents to preventing them entirely.


Conclusion

The presentation at the Queensland Mining Industry Safety and Health Conference highlighted the transformative potential of Generative AI in incident investigations. While the technology is not without its challenges, the ability to reduce human bias, automate time-consuming tasks, and standardize investigations is a significant step forward in ensuring workplace safety. We left the audience with a clear understanding of both the current capabilities and the exciting future potential of AI in the mining industry.


We’re eager to continue exploring how AI can further enhance safety practices and drive the mining industry toward a safer, more reliable future.

14 views0 comments

Recent Posts

See All

Comments


bottom of page