While I have few ideas which are more of a short term and for me to learn the basics, I have following ideas in my mind. I need to admit that I was sceptical about posting it even before first version of PoC is ready. The risk of some body implementing it before me exists ( I am over optimistic to be precise, but you never know). However I see that by publishing about my plans
- I can get honest feed back.
- Help, Tips,Suggestions to make it better. Perhaps people better than me can collaborate with me.
Following are the projects. Feel free to contact me if you feel if the ideas are good and worthwhile pursing it and if you would like to collaborate/help me.
Goal is to develop a IoT platform/solution which can collate data from vehicles and can apply intelligence and analytics on top of the data to decipher their real world usage, which can help in decision making.
There are different parameters which can measure the quality of how vehicles are driven. Few aspects are sudden acceleration, deceleration, change in directions (zig-zag ride), unusual heat of Engine , mileage, and measurable data about other components of a automobile (bike,car,bus etc). If this data along with environmental data are collected and stored, algoirthms can be applied against it to determine how well or how bad a vehicle has been driven.
Each component, which can give measurable data associated to vehicles,usage or drive quality will be associated with a sensor. Sensors collect the data and send it to a Data collection service. When a sensor unit (which combines set of senors for the automobile)is installed in a vehicle ,it will be uniquely associated to a vehicle (e.g Engine serial number). When a customer purchases and activates the sensor unit, it will register the vehicle, driver and sensors data with the common service. Sensor unit will also be paired with a smart phone and an app will be available (Incase of cars, smart phone might not be required and cars infotainment system can be used). When a vehicle is driven and data is gathered, it will be stored in a temporary storage and will be pushed to intermediate hub (app or infotainment system). From here, data will be pushed to common service offline whenever user is connected to internet. All communication i.e between sensor unit to app and from app to common service will be encrypted and cannot be decrypted. So there is no chance of users to tamper with the data.
Analysis using big data frameworks can be done on the data collected for a vehicle to measure the overall quality of how vehicle is driven over a period of time and at various points of time.
Below are few use cases or real world problem, the solution can address.
- Used Vehicle Driving Pattern for Used vehicle purchase
Use case/ Purpose: To make used vehicle purchase more objective and Data driven.
How it benefits the customers: Vehilce owners, or prospects can view the data and get more insights into real behaviours or the way vehicle was subjected to. Mechanic need not be the only one to give certificate. Prospects can also get real data about mileage of the vehicle etc. This will help them to choose the one among various options.
Business Model: Similar to KBB or Credit scores, customers need to pay to get access to data. Vehicle owners can pay one time charge for installation.
Challenges: What if Sensors are damaged in accident or a new spare parts are obtained. How do we analyze and give score for new parts.
2. Driving pattern Driven Insurance Premium
Use case/Purpose: To help insurance companies map the risk of drivers and determine premium more accurately.
How it Works: Sensors connected to the vehicle ( 2 wheelers,4 wheelers), will collect data which can give idea about driving quality. For e.g sudden acceleration, deceleration, temperature of the engine, quick turns made,how many collateral hits avoided etc. The data will be mapped to a driver as well. Insurance companies can get data about the vehicle and driver and identify how good the vehicle was driven and how good the drivers in the policy are. With this they can find the risk profile. This will be more accurate than carpet risk profiling based on gender,age,and vehicle type.
How it benefits the customers: Customers who drive well and mostly safely can be classified as a less risk category and be provided with a lesser premium. Similarly rash drivers can be charged a premium.
Business Model: Similar to KBB or Credit scores,customers and insurance companies need to pay to get report of the risk profile of the vehicle and driver.
Challenges: Driving pattern varies based on the context. Also if a vehilce is driven by multiple people, how do we map it to specific drivers at different time in a seamless manner.?
3. Safety Rating for Buses/Routes
Use case/Purpose : To help passangers choose the safest transport provider for overnight bus journeys.
In India 100s of road accidents occur per year involving public transport vehicles. This includes buses driven by private transport operators, government Transport Corporation operated buses and buses booked for private events. Though the number of accidents involving these buses might sound small, the physical and emotional trauma it causes is immeasurable and the number of accidents that are avoided in last minute could be much higher. Main reason for the accident is over speeding, reckless driving by the drivers. This platform can provide a solution for it.
How it Works: The sensors could measure different statistics to measure quality of driving. Reporting by the service could measure the safety of the fleet provided by Transport Corporation or a specific route. Customers can have access to the level of safe driving of each and every public buses. With this customers can choose the operator who assures more safe and careful driving than a operator who ask their drivers to drive in a reckless manner.
4. Emergency Stop based on Drivers health
Sensors attached to important parts (break, Fuel injection) can be attached to smart watches and wearable of the driver. When a driver experiences severe health condition like heart attack or faint, the sensors can force the vehicle to stop or reduce the speed or can alarm the passengers.
5. Realtime Pro-active service readiness
Another variation of the service can give best experience to the drivers. Sensors attached to the parts of vehicle can determine when the part malfunctions and if the vehicle cannot run anymore (e.g Tyre puncture or Engine overheat etc). The sensors will communicate with app and the app will look up the net for nearby (based on location) workshop or mechanic who can solve the problem and will provide to the drivers. Drivers can just call the mechanic to get it fixed. Or a spare can be ordered well before hand and when they arrive mechanic would have it ready and he just need to fix it.