ICBC - Telematics for New Drivers Pilot

ICBC LogoDo you have less than 5 years of driving experience and hold a BC Driver's licence? If you are interested in taking part in a driver telematics pilot project ICBC is accepting up to 7,000 volunteers to see if using telematics will make you a safer driver.

 

Drivers that participate will be eligible to receive rewards for making good driving decisions. If you are interested, visit the registration page.

The telematics software, installed on your cell phone, combined with a small in vehicle device will monitor speed, braking, acceleration, turning and level of distracted driving. Participants will receive feedback on their driving that will allow them to improve.

This step is an expansion of a pilot conducted earlier this year that prevented participants from using their cell phone while driving.

"From our first telematics pilot earlier this year, ICBC has developed a telematics strategy to identify how the technology can be used to improve road safety and drive behavioural change among higher-risk drivers in B.C.," said Nicolas Jimenez, ICBC's president and CEO. "We heard from those pilot participants that most believed the use of telematics would make the roads safer for everyone. This is our next step in a thoughtful examination of telematics technology and how it might help to keep these drivers safer."

This will undoubtedly pave the way for insurance rates based on your driver risk as shown by continuous monitoring. Belaire Direct's Automerit is an example of this already in operation here in BC.

There are also free apps such as DrivSafe that a curious driver can use to improve. Others, such as edriving's Mentor will soon be available for a monthly fee and will include coaching.

Having an app monitor a drivers' every move is an interesting approach.  In order for speed monitoring to be relevant, there must be corresponding gps tracking, otherwise the speed limit on a given section of highway cannot be compared to the vehicle's speed, making this piece of data meaningless.  As a driver, I would view this as an ivasion of privacy; the app would be tracking my vehicle's, and therefore my own whereabouts as long as it is activated.  This location data would then be stored on a server for the use of...who?  And what about security?  Even the CRA can't keep our tax data safe in recent years.  This is a serious concern.

... as Alan Parsons sang

Indeed, a serious concern. Big Brother, 2018 style.

I wouldn't download something like that to my phone, with no idea what they're going to do with the data, and no way of controlling it. And frankly, ICBC are the last organization I would trust to have this information. A person would have to be a fool to sign up for this; or perhaps just an ingenuous youngster who already allows far too much access into their lives already by potentially unfriendly forces. 

I was in favour of driver monitoring until I participated in the ICBC trial. Althoiugh I achieved a very high score overall, I found that the system was not a very good judge of safe driving, as it penalised some actions it should have not, and failed to penalise other actions that it should have. Here are some of the problems I had:

1. No demerits issued for speeding when I deliberately exceeded the speed limit by 10km/hr.

2. I don't believe it issued demerits for other illegal actions that could increase collison risk, e.g. left ot right turning where prohibited, wrong way driving on the street, failing to yield on crosswalks and elsewhere, etc.

3. When approaching a traffic light that had just turned to amber, I followed the rule to stop if it's safe to do so, i.e. when before the point of no return, but sometimes received demerits for braking a little harder than my usual very gentle braking. If I had broken the law and continued through the intersection when I could have safely stopped I would not get the demerit. In other words the safer legal choice was penalised.

4. When conditions allow and in the absence of potential hazards, a typical battery-electric car will safely accelerate up to the speed limit more rapidly than a typical gasoline-powered car, but the system did not appear to take this into account and would sometimes issue demerits for too rapid acceleration in circumstances where the rate of acceleration was not at all excessive or dangerous.

5. No indication is given as to when the vehicle is approaching or has reached the demerit point for braking, turning or accelerating. I am not suggesting a noisy alarm should go off but if the intent is to educate, the driver should know at the time what the system doesn't like rather than having to review the log later, which probably isn't going to happen.

6. There is no appeal of the score, which will result in some drivers paying more for insurance than they should.

It appears that a major problem with the system is because the algorithms used to judge behaviour did not take into account all the factors that for example a driving tester would. It reminds me of the biased HR programs that are used to review job application resume's.

I would be in favour of the concept if accuracy and privacy could be assured and the scores could be appealed, but based on the trial I don't believe it's ready. Computer systems for autonomous cars are still having a lot of trouble ensuring that the degree of safety of the cars they control is properly assessed and ensured. I believe the driver monitoring app is in a similar state.

 

This is not a system to protect the interests of the drivers/insurance holders or for making the roads "safer".

At the core, it is simply a system to extract more money from car insurance holders, while providing a "plausible" reason as to why the insurance holder is being ripped off.

If everyone gets an opt-in discount - it means that everyone who opted-out is getting ripped off.
Everyone who opts-in can lose their discount and be charged a surcharge - because they are now a "bad driver" based off monitoring.
None of the data points that the system monitors are directly related to the risk they present to their insurance company.

The only valid individual-specific measure of insurance risk is their prior claims history, and maybe the amount of trouble-free miles they have traveled. Neither the speed, nor the lateral forces generated by turning, braking or accelerating have any direct correlation to how much or how little damage a specific individual may inflict on others.

At-most, these systems could be useful for authenticating the driver to the insurer for each trip in order to record the miles each individual driver has traveled, so that the discount can be applied per X-number of trouble-free miles, instead of how it is applied now - 5% for simply holding a valid driver license each year. Although that is also not in the interests of the majority of drivers/insurance holders, as people would have to actually drive (action) to acquire the discount that would otherwise be acquired by the passage of time (inaction).