Data to save lives on the road
By Sharala Axryd & Chari TVT May 23, 2019
- 80% of accident in Malaysia are caused by human error not road and vehicle conditions
- Telematics has been proven to change driver behaviour for the better
IT KILLS more people than cancer globally and in Malaysia. If there’s one thing we are overachievers in, it’s road accidents. Every one minute, an accident happens on our roads and every hour or so, another life is claimed and lost forever.
In 2016, Malaysia ranked 15th in the World Bank Group’s Doing Business Report, 22nd in the World’s Talent Ranking and third when it comes to road accident deaths worldwide, surpassing the world’s two most populous countries; China and India. In the same year, the number of annual road deaths recorded at an all-time high of 7,152 official cases.
The aforementioned third placing in road accident deaths worldwide was attributed to people driving recklessly or ignoring traffic rules, instead of road and vehicle conditions, according to research by the Malaysian Institute Road Safety Research (Miros) which also highlighted that 80% of accidents are caused by human error.
Despite a world-class highway system on par with developed countries, Malaysians have a specific driving culture and mindset when they are on the road, especially when it’s congested. Frustration, road aggression or even drowsiness while sitting behind the wheel, selfishly hogging the lane, incessant weaving, dangerously forcing their way, switching lanes at the last minute to ‘cut queue’, these should all sound familiar.
Couple this with the fact that according to Nielsen’s Global Survey of Automotive Demand, Malaysia is the third highest in terms of car ownership in Southeast Asia at 93%, and you get a lot of reckless drivers on the road.
Malaysia’s driving exams also don’t put enough emphasis on keeping a safe distance behind cars, which in turn affects speeding with changes in braking and stopping. This is significant, given the fact that based on work by Nilsson in Sweden, a change in average speed of 1 km/h will result in a change in accident numbers ranging between 2% for a 120 km/h road and 4% for a 50 km/h road.
More recently, PLUS Expressway Bhd cited illegally parked lorries and trailers in emergency lanes as one of the main causes of accidents on the expressway. PLUS has reduced the number of illegally parked heavy vehicle-related accidents since 2017 through heavy monitoring with the collaboration of the police and Road Transport Department (RTD). Other measures taken include the RTD taking over the Automated Enforcement System (AES) cameras as part of Automated Awareness Safety System (AWAS).
In comparison, Monaco, which is famous for its Grand Pix, is in possession of the lowest traffic mortality rates at 0 for every 100,000 people. The public transport system in Monaco is heavily monitored where proposed transport infrastructure is evaluated to ensure they meet safety standards for both pedestrians and vehicle operators. Drunk-driving and helmet laws are enforced and have been effective in preventing accidents while the urban speed limit is at 70km/h to prevent speeding.
Japan has also taken preventive measures which has resulted in traffic fatalities totaling 3,532 in 2018, a decrease of 162 compared to the previous year and a record low since statistics were first kept in 1948. With high pedestrian and cyclist deaths compared with the United States and Europe, Japan implemented several measures to protect pedestrians such as introducing ‘Zone 30’ areas where the maximum speed is limited to 30 kilometers per hour. Side strips are established or widened and there are barrier-free traffic signal equipment for the elderly and mobility-challenged. Also effective is a recent move to equip vehicles with safety technologies such as collision avoidance systems and lane departure prevention systems.
It is a known fact that road accidents directly affect both the environment and the community. However, it also has a substantial impact on a country’s economy. A report by the World Bank in 2018 illustrated that road deaths and injuries sustained from automobile accidents hinder economic growth in developing countries.
According to a study conducted by Universiti Putra Malaysia (UPM) on the economic aspect of road accidents, the cost of a life lost as a result of a road accident works out to be RM2.3 million inclusive of medical and insurance costs. Hence, a developing country such as Malaysia should efficiently allocate its resources to decrease its accident and road fatality rates in order to achieve impactful economic progress.
Data by Miros
Even assuming insurance companies pay RM3,000 per accident on an average, total insurance settlements every year will be a staggering RM1.5 billion. Insurance companies generally increase the premium of everyone rather than reducing premiums of well-behaved drivers and substantially increasing the premium for bad drivers. Insurance companies need data for good and bad behaviour to differentiate the premiums charged to their customers.
For a nation that is aspiring to become a developed country, Malaysia should look towards decreasing its traffic-related death rate – World Health Organisation
GPS data for actionable insights
While apps like Google Maps, and Apple Maps dictate directions while taking into account time, traffic and tolls, Waze’s crowdsourced, user-centric approach includes user-reported accidents, traffic jams and other road conditions in real-time.
This open data source allows for the understanding of traffic flow issues to create traffic management systems, support traffic flow management decision-making and identify traffic accident hotspots.
By analysing real-time and historical city data, several use cases have determined instances within Malaysia, specifically the Johor-Singapore causeway and Klang Valley, where traffic is most severe by the time and date, bottlenecks and hotspots for accidents as well as congestion patterns throughout seasons.
This information is based on use cases conducted by The Center of Applied Data Science Data Science team. Our data scientists collaborated with third-parties to collect Malaysian traffic-related data to create actionable insights through advanced analytics.
Are telematics a panacea?
So, what is the solution to Malaysia’s road safety issues, accidents and fatalities? The answer lies in telematics, which has been proven to change driver behaviour for the better.
What exactly is telematics? According to Gartner Inc, telematics is, “the use of wireless devices and ‘black box’ technologies to transmit data in real time back to an organisation. Typically, it’s used in the context of automobiles, where installed or after-factory boxes collect and transmit data on vehicle use, maintenance requirements or automotive servicing.”
Telematics is the amalgamation of tele and matics: abbreviations for “telecommunications” and “informatics”. A telematics system collects data and information about vehicles which it then sends, receives, processes and stores via telecommunication channels such as GPS and cellular devices.
Strengthening the provisions for regulating road users’ behaviors will contribute to the reduction of road traffic deaths and injuries in India”, said David Cliff, CEO of the Global Road Safety Partnership, while focusing on the specific issue of ‘bad infrastructure vs bad driver’.
Wondering how effective the use of telematics is in automobiles? Telematics systems boast the capability to record data of varied driving habits which include components such as average speed, acceleration, braking tendencies, speeds at which turns are taken, fuel usage, idle times and location. Armed with this information, improvements to driving habits can be made.
Let’s take a look at a simple case example where telematics has come in handy to not only boost productivity and reduce cost, but also reduce accidents. In the fleet management of trucking logistics, employers use telematics to remotely identify potentially negative events in their trucks such as hard braking, excessive acceleration, poor route planning or insufficient tyre pressure. They use this data to coach their drivers into improving their performance.
The Malaysian Transport Ministry is in the midst of strongly considering telematics devices to assess driving behaviour through a nationwide driver profiling system. MiroS postulates that by scoring drivers based on real-life driving behaviour, it can motivate driver-improvement as well as indicate both good and risky drivers.
The initiative is one of many developed under the Transport Ministry’s impending reward programme concept for drivers to claim rebates on fuel or toll fees based on their road behaviour. An organisation is also currently developing a mobile app to monitor and evaluate road manners to further bolster this initiative.
Protecting drivers during a crash through the use of seat belts and airbags is without a doubt, compulsory for safety. However, the focus is shifting from protecting people in a crash to preventing that crash in the first place.
So how does technology fit into this? Let’s take a look at some of the examples below:
- PDRM to mandate inserting an IoT device ( SIM card, smart camera etc) in every transport vehicle be it bus, lorry, motorbike, train, cars both private and enterprise and track driver/rider’s behaviour ( good, bad, ugly) using telematics software. Pass on the information to regulators, insurance companies and others for fixing premium and settlement. Use the data to manage traffic flow.
- Introduce carrot and stick insurance policies for good and bad behaviour. For example, an insurance company will pay for third-party claims in cases of drunk driving but will not reimburse for self or vehicles.
- Using the same IoT device, automatically shut off engines specially for motorbike riders if they exceed certain speed limits (technically this is possible today).
- Use drones to track speedsters and warn them or cut their speed.
- Use the IoT device to detect drunken driving (facial recognition, smell detection, detecting tiredness of the drivers – specially bus or lorry drivers)
- Video camera (IoT) in each car to detect the use of mobile phones while driving and impose punitive measures such as increasing insurance premiums substantially or taking away the driver’s licence.
- Flat tyres and bald tyres are major causes of accidents. IoT device can detect and tell owners of these issues. The insurance coverage will be removed or premium raised if the tyres are not replaced . Alternatively make vehicle inspection compulsory once in every two years
- Electronic stability control (ESC): This system does a good job of preventing cars from sliding out of control, by applying the brakes momentarily to individual wheels and/or reducing engine power.
- Automatic emergency braking (AEB) : This enables the car to automatically brake in an attempt to avoid a collision.
- Lane departure warning and prevention: A departure warning will alert the driver if the car is drifting out of its lane, while lane departure prevention will try to steer a vehicle back into its lane.
- Blind spot warning: If there's a vehicle sitting in your blind spot, a monitoring system will alert you with a warning light inside or near your side mirrors.
- Front/Rear cross traffic alert: These systems can alert the driver – when backing out of a parking space or crossing an intersection, for example – if there's a vehicle approaching from the side.
Installing systems like these, or making it a requirement that they be included in vehicles, will help reduce the rates of crashes; especially those caused by driver error.
Analytics is also another method to assist in decreasing the number of accidents. Predictive analytics use many techniques from data mining, statistics, modelling, machine learning, and artificial intelligence to analyse current data and make predictions about the future.
By using predictive analytic software, law enforcement can actually predict when accidents will happen before they occur. Take The Tennessee Highway Patrol for example; they are one of the few law enforcement teams out there who have started using a programme called Crash Reduction Analysing Statistical History (CRASH), which analyses loads of data that includes weather patterns, crash history, football schedules, when special events happen, and even the location of places that sell alcohol in order to determine where the most likeliest places for crashes are and when they might happen.
Once the locations are determined, patrols will be deployed to those points as a means to control traffic conditions and keep everyone driving safely. In an event that an accident occurs, assistance can be offered almost immediately.
Achieving the goal of accident prevention with technology and analytics
Is it possible? It could be.
Signs that can detect and alert drivers of impending dangers, for example, a series of large red lights mounted above the roadway, which begin to flash when a red light runner is detected, to warn motorists with the right of way to stop and avoid a crash.
Using sensors embedded in the roadway, a large blinking "No Left Turn" sign lights up when vehicles in the crossroad are too near or travelling too fast for a safe left turn.
An in-vehicle system that works with the sensors embedded in the roadway triggers a high-pitched beep and a red light on the dashboard when a left turn is unsafe.
Other type of smart signs can include mobile warning road-signs, which aim to highlight the dangers of driving while using a mobile phone, flash a warning sign to motorists whose car is transmitting a usage signal. The sensor will pick up that a driver is using a phone for calling, text or data purposes and will activate the warning sign.
What if all automobiles come equipped with safety features rather than having them as an option during the purchase? Te features would include ESC, AEB, lane change alert, blind spot warning and their like. If all automobiles had these systems installed, then the next step of prevention would be easier.
By implementing a point-based system, drivers will have to register through an app that allows road enforcement officers to monitor their driving habits. From here on, the system will work in the same way as when you get a ticket for breaking road safety rules. Good drivers can receive rewards such as discounts on fuel or insurance, while those with bad habits can be summoned to attend safe driving courses.
With drivers earning points for safe driving, and losing them when they engage in distracted or unsafe driving habits, awareness will be instilled in them to practice safer driving. By using gamification methods — backed by telematics data, which takes the objectivity out of the equation — drivers are encouraged to make better decisions over the long term, which in time become good habits they will take with them every time they get behind the wheel.
Malaysian drivers also need to install apps that can be an SOS as a prevention measure while also easing the unpleasant process of reporting collision incidents while in a rattled or injured state. Don’t underestimate communication technologies and social media tools as a solution to emergency responses, connecting with rescue specialists and maintaining personal safety.
Demand and development of saviour mobile apps skyrocketed when Facebook activated the Safety Check tool in response to the Paris attack in 2015 that promoted users in affected areas to confirm their safety.
Rather than rely on a “spray and pray” approach when it comes to traffic safety, technology and data analytics could help better target the Malaysian government’s efforts by understanding where and why accidents occur.
Through informed data-driven decisions on common causes of nationwide traffic trends and accidents, government agencies could then make smarter, safety-conscious investments in the country’s infrastructure.
Investment in analytics and machine learning technologies can lead to more nuanced and effective safety policies by offering a holistic view of traffic safety.
Sharala Axryd is the founder and CEO of The Center of Applied Data Science (CADS) and Chari TVT, Consultant & Board of Advisory, CADS