Avalanche safety offers lessons to prevent tailings dam failures
Tailings safety can be improved by learning from other disciplines and by acknowledging human factors
Avalanche safety may sound like it has nothing to do with mine waste, yet it is a discipline where the impact of human factors on failures have been researched, discussed and published – and so offers an opportunity to learn! Year in, year out, experienced and qualified groups succumb to avalanche hazards, despite advances in equipment and technology and, sometimes tragically, these groups learn lessons the hard way. What leads to the decisions these groups make, that ultimately turn out to be poor decisions? Can we avoid making the same mistake in tailings and waste management, when it comes to making risk-based decisions, by paying attention to human factors or the way we assess evidence and make risk-based decisions?.
If nothing bad happens, we may evaluate our decisions as the right ones—not necessarily because those decisions were good, but because we didn’t receive any feedback that they were bad. As such, we may inadvertently form inaccurate assessments or develop potentially dangerous habits based on poor decision-making
After the death of an expert friend in an avalanche incident, a researcher named McCammon investigated over 715 incidents –his research on human factors changed the way avalanche workers think about avalanche incidents. It provided insight into the key question ‘how do people come to believe that a slope is safe, even when they are faced with likely evidence that it isn’t?’
McCammon introduced the concept of heuristic traps. Heuristics are simple, efficient rules, learned or instilled by evolutionary processes, that explain how people make decisions, come to judgements, and solve problems typically when facing complex problems or incomplete information. These rules work well under most circumstances, but can lead to errors or biases, especially in unpredictable, high-risk environments.
So, heuristic traps are mental shortcuts that can result in common decision-making flaws. Like avalanche terrain, tailings operations can also be poor feedback environments, as there is no immediate feedback on the decision-making process. This means we are tempted to mistake good luck for good decision making. Some heuristic traps, which are relevant for tailings safety are:
- Familiarity/complacency, where we believe a course of action is correct because it has been done before without incident.
- Social acceptance, where we fail to speak up about uneasy situations.
- Commitment where we are driven by a goal, and become blinded by it. For tailings this might be raises, mine production etc.
- Expert halo, where people perceived as experts can dominate the decision making, even if they step outside their realms of expertise (an issue at the Oroville Dam failure),or are not actually expert. This is important for tailings safety as we rely on experts and we are challenged to find enough experts.
- Social facilitation, where we believe our actions are correct because other people are doing them.
- Evidence review biases, where we can gather facts that support certain conclusions and disregard facts that threaten them. We can also be conservative and fail to change opinions in the light of new evidence or we allow recent or more frequent events to dominate thinking.
- Illusory correlations, where there is a belief that patterns are evident or variables related, when they aren’t.
- Selective perception, where people see problems in terms of their own background and experience.
- Underestimating uncertainty, where we may be overconfident about what we think we know.
Be heard and not herd
Avalanche research has shown that people do not always make the right decisions, based on human factors. As engineers are human too, acknowledging human factors and implementing these ideas may improve tailings safety:
- Improve communication and grow teams that are curious, questioning and encourage speaking up. I have heard this referred to as ‘heathy tension’.
- Create diverse teams, that aren’t reliant on the decisions of one person, or on very large groups. Small group decisions are usually better and diverse technical teams minimize selective perception.
- Be wary of using judgement based purely on experience, which may harbor evidence review biases.
- Consider and develop Plan Bs, to help reduce the commitment pressure.
- Use sensitivity analyses to indicate how changes in assumptions may impact results and projects.
- Share lessons and publish near misses, to help prevent poor feedback environments.
At Stantec, we understand that human factors can influence tailings safety and we build teams, systems and expertise that acknowledge this; this allows us to balance tailings risk and provide solutions that create thoughtful and effective governance.
Lucy Philip, a Principal and Senior Geotechnical Engineer at Stantec Calgary, is presenting further discussion and ideas on ‘Human Factors: Can we learn from avalanche incidents and apply the findings to tailings and mine waste?’ at Tailings and Mine Waste 2019 in Vancouver. You can see her presentation on Monday 18th November at 2pm in the Management Theme Session 3.