You’re just out of the office and want to take a bike or scooter home. You open your favourite app and see there’s one available at the end of the road. Perfect, but not quite. You get there and it’s unusable. Why does it show up on your app then?

You walk a bit further and find another one, problem solved. Wait. You start your journey only to find out that the power keeps cutting out on the scooter. You might think, “that’s the last time I use this service”.  

Unusable vehicles in a fleet are a cause of massive frustration for users, and can have a dramatic effect on an operator’s revenue and, therefore, on profitability. Self-diagnosis features are the key to solving this massive micromobility challenge from both a customer-facing and an operational perspective.

The challenges of undiscovered vehicle faults

Safety first

Every operator is concerned about their users’ safety. To prevent accidents and injuries, it is paramount to catch these issues before they become real problems. They need to be addressed before a rider begins a journey so that they are never at risk.

More trips, more revenue

Simply put, the higher the fleet availability, the higher the revenue. When you have hardware on the streets that isn’t being used, it’s missed revenue.

User retention

Less obvious, however, is the knock on effect that it can have on user retention. If a user tries to make a trip, approaches your vehicle and sees that it is damaged, they are going to go away frustrated. You also can’t always count on users to report the issue so, in a worst case scenario where the vehicle is parked in a busy area or it’s rush hour, you could be looking at 4 or 5 users encountering the same problem in just one hour.

Users can forgive this, of course. What is perhaps more problematic is that when a user decides not to take your vehicle, they use another operator’s service. If they already have their app, their decision will be instantaneous.

When this occurs multiple times in a week or month, you could lose that user forever.

Self-diagnosis features solve the challenge

From an operational perspective, the challenge is identifying problems and solving them before a user has a bad experience.

The new era of connected electric bikes and scooters solve this challenge. They are equipped with embedded sensors that monitor mechanical, electrical, and software systems to proactively detect potential failures before they become problems.

When an issue is detected, one of two things can happen:

1. Possible fault: The information is transmitted to the back end in real time, alerting the operations team that they might need to make a decision on a potential problem. The vehicle stays online because it’s not a critical fault that will impact a user’s experience or safety. (e.g. battery temperature lower than usual). It's then up to the ops team to decide whether to send a field agent or not (and at what priority).
2. Agent required: The sensors detect a problem that requires human attention to fix. The bike takes itself offline and communicates with the backend via 4G, creating a ticket with the nature of the problem so that a field agent can be deployed (e.g. high chain tension).

Typical use case - solving pedal assist inefficiency

It goes without saying that one of the most important aspects of an electric bike is that the pedal assist feature runs efficiently. One of the things we are able to do at Fifteen is detect that the chain tension is too high, which impacts the assistance and makes for sub-optimal performance.

Depending on how the operations system is set up, the vehicle can either stay in service and continue to monitor the tension, or take itself offline and flag itself to be checked by a field technician immediately.

This is a huge benefit because not only does it ensure riders have great experiences, but it also allows the bikes to be seen before the chain tension leads to a worse fault that stops the bike from being used completely.

Other examples

• Battery - does the battery show signs of physical or water damage? Is it becoming too hot or cold to operate effectively?
• Mechanical - reporting on the status and effectiveness of brakes, motor etc.
• Connectivity - are there connection interruptions to BLE, WiFi or LTE-M?
• Software - has battery consumption increased since the latest firmware update? Electrical - is the bike unlocking and locking properly, and is it transferring power to the other bikes in the station?

Using all the data you can also identify what problems you are likely to face in the future and plan accordingly. Perhaps this means you adjust the number of field agents you have out on any one day, or change their locations or priorities.

One thing is certain, this data will help you maintain high fleet availability, increase your rides and, therefore, increase revenue. The self-diagnosis features will also cut your operational costs, because field agents will take to the streets already knowing what the problems are, equipped with the right tools.

Optimizing field operations

Self-diagnosis signals play a crucial role in optimizing maintenance operations during the day, and even the following day. If field agents know what the problems are without having to work them out on site, they can come equipped to the location with the right tools and fix the issue with minimal fuss. This saves a considerable amount of time.

It also means that the team knows, before they even start their round, which vehicles are quick-fixes and which ones will need to go back to the warehouse. Vehicles with minor problems can be prioritized, contributing to increased fleet availability and increasing daily ridership.

The low-down

• Detecting potential issues before they become a reality has a significant impact on user experience.
• Knowing what to solve up-front allows field agents to be better organized and optimize their routes - which can also contribute to improved workforce management and lower operational costs.
• More fixed vehicles means more ridership, contributing simultaneously to user acquisition and retention!