FATIGUE MANAGEMENT
Orica virtual co-pilot system
Keeping drivers and operators safe in mining is not restricted to large rigid haul trucks . In FY2022 , global explosives and blasting systems major Orica conducted a driver assistance trial in Latin America to identify new ways to improve vehicle safety . Rita Carvalho , Head of SHES at Orica LATAM told IM that the trial comprised a ‘ virtual co-pilot ’ system that controls and monitors the speed and velocity of vehicles from external transport carriers delivering Orica products and provides real-time information on the risks and hazards of the defined route to drivers .
The system incorporates a digital Risk Route Assessment ( RRA ) that records the speed and velocity of the vehicle per route section and provides real-time information to the driver and a supervisor of the transport company whenever a vehicle exceeds the system ’ s recommended speed limit , parks at an unapproved location or when the driver has insufficient rest during the route .
Carvalho said : “ Safety is Orica ’ s number one priority . These measures ensure the safety of drivers by preventing injury and fatality , and the security of Orica ’ s products as they are being transported . Safe and timely delivery of our products are outcomes for our customers . We are also working with our contractors to provide training and awareness sessions onour requirements and the fundamentals of vehicle stability , as well as encourage immediate feedback about driver behaviour and performance .”
An awareness training related to vehicle stability called ‘ Slow down on the route ’ was developed and distributed across the region , with a focus on Orica MMU™ ( mobile manufacturing unit ) drivers and external carrier drivers to minimise the risk of rollover events . Modules that formed the training included : Speeding and how to navigate a curve safely ; centre of gravity and weight and centrifugal and centripetal force ; plus rollover contributing factors and liquid load and case study . Every MMU™ driver and external driver who provides services to Orica in LATAM has to complete and pass the training .
Designed to shape driver behaviour is another program called ‘ Sharp eyes on the road ’ that has been rolled out in Brazil . A pilot was launched in Peru with external transport carriers being invited to participate in two workshop categories . Firstly for supervisors to introduce and explain the mechanics of the program and how to implement it . Secondly for drivers , where the 10 most frequent causes of events on routes were identified and fed into a checklist .
devices also capture sleep data for even more personalised fatigue predictions . All sleep data are kept strictly confidential and are not shared with supervisors .
IM spoke to Robert Higdon , SVP of Product and Marketing at Fatigue Science who said : “ The single biggest development at Fatigue Science over the past year is our technology evolution , which now allows us to predict fatigue without requiring a wearable . On this journey , we ’ ve gone from requiring a wearable to be worn 24 / 7 , to only requiring it to be worn while sleeping , to only requiring it some of the time and we will fill
ReadiSupervise Mobile from Fatigue Science helps mine shift supervisors more effectively manage fatigue risk , improve operator safety , and increase crew performance in daily operations in the gaps . Today , we ’ ve come full circle : customers don ’ t need to use wearables at all . Now , we can provide reliable fatigue predictions purely via software .”
Higdon says this has been transformative because F / S can now go to mine sites that were reluctant to adopt wearables and offer a solution that fits their needs . So why were some mines hesitant to use wearables ? “ This actually varies region by region – for example , some North American firms cited privacy concerns , while some firms in South Africa had worries about the potential for devices to be lost or stolen .”
For mines that have known F / S as a wearable-based system , the challenge has been to educate them on recent advancements . “ We show them how we are able to leverage a pool of four million anonymised , de-identified sleeps from other wearables to give them the benefit of not having to deploy wearables themselves .”
Sleep data were recorded with validated wearable devices at mine sites around the world , including the ReadiBand , ReadiWatch , and devices from Fitbit and Garmin . “ For mines coming to us for the first time ,
knowing our scientific background and machine learning capabilities , it ’ s a logical next step . Biomathematical modelling has been a trusted science long before wearables emerged , and now , with ML and a vastly larger pool of data , our automated system can provide daily insights without eh manual processes of years past .”
So how does the non-wearable approach work ? “ While it is ‘ zero wearables ’ from day one , we do initially need other data from operators . That includes height , weight , BMI , age , sex and a one-time questionnaire on factors like sleep environment – is it too hot or too cold – commute , diet , and more . It also includes some data about the mine environment – such as sunrise and sunset times – which affect circadian rhythms in a highly predictable way . There are about 30 questions that take about five minutes to complete and data remains private ; from there , machine learning forms a personalised sleep profile for each individual in combination with their day to day schedule . It leverages data from similar individuals in our wearable data set to estimate how each individual without a wearable might be affected by a particular work schedule . Our machine model then makes day to day hyper-personalised estimates of the sleep that each worker had , based on their schedule , which indicates when they had the chance to sleep , in combination with their sleep profile .”
Clearly there are going to be some anomalies , such as where a person missed more sleep than usual due to a young baby crying during the night , or there was a storm that kept them awake . Wearables will account better for these . Higdon says Fatigue Science is upfront with mines that if those edge cases are the level of precision they are looking for , then it can offer a hybrid approach that includes wearables for some or all workers . “ But most of the time , mines are looking to go from 0 to 1 . They don ’ t have any insight into any of the causes of fatigue or where their worker fatigue stands . Most of the time it is a combination of the schedule and the person ’ s personal sleep profile that causes fatigue . “ Someone with a particular sleep profile might do well on day shifts but might really struggle on the fifth night shift , as one example of how complex interactions might play out in the machine learning model .”
Do customers who already have a reactive solution ( such as cameras ) in place to detect microsleeps also have interest in a predictive solution , such as Readi ? Higdon stated : “ Many of our customers come to us having already deployed one of the camera-based solutions . The two approaches are definitely complementary . On their own , the camera systems sometimes give some false alarms , which can disrupt productive operations . Customers want the ability to take some kind of preventative measure at the
72 International Mining | JANUARY 2023