Monetizing Big Data in Fleet-Based On-Demand Shared Mobility

In my book and previous posts I build a broad case for the key role big data and AI play in next-generation mobility, and provide several examples from transportation and logistics. Next-generation mobility is about intelligent, connected vehicles that utilize some form of electrified propulsion, and on-demand shared transport services of people and goods that will be offered through such vehicles. Many of these vehicles will be capable of autonomous movement. Next-generation mobility will help us address some of our biggest challenges, such as pollution and climate change, urbanization and congestion, aging population, and traffic fatalities, while enabling us to maintain economic prosperity by operating highly optimized supply chains that span the globe. It will give rise to a new value chain where big data and AI will play a key role. It is therefore important to identify the new monetization opportunities enabled by big data and AI in the context of this value chain.

Introduction

Despite the growing number of announcements about technological achievements, investments, partnerships and acquisitions relating to autonomous vehicles utilizing Level 4 or Level 5 driving automation, we are still several (5-10) years away from the broad utilization of such vehicles, particularly for passenger transportation. Many incumbent automotive OEMs envision that Level 4 and Level 5 autonomous vehicles that will be used for passenger transportation will be privately owned.  However, as we will see in the next section, the majority of the use cases being tested today involve fleets of autonomous vehicles. For this reason, in this post I first examine the value that is added by big data and AI in every component of the fleet-based on-demand shared mobility value chain that I have previously defined. I conclude by identifying the monetization opportunities that big data and AI offer.

Six Use Cases For Next-Generation Mobility Vehicles

Based on my research I have identified six general use cases where autonomous vehicles can have transformative impact. The successful deployment of these use cases will depend on a number of different but highly interdependent factors that include:

  • Creating the right technologies at reasonable costs and energy efficiency to enable autonomous vehicles to deal with environments of increasing complexity that can change dynamically;
  • Devising scalable business models;
  • Building the appropriate transportation and electrification infrastructures to accommodate the growing numbers of autonomous vehicles;
  • Instituting the appropriate regulations;
  • Resolving liability issues;
  • Addressing issues relating to cybersecurity and data privacy, and
  • Achieving social acceptance of autonomous mobility, including the resolution of labor issues that will undoubtedly arise.

I have ordered the six use cases based on increasing complexity:

  1. Specialized vehicles operating in controlled environments. Corporations have already started deploying specialized autonomous vehicles, some of which may be electrified, in environments where employee injury risk is high and labor shortages exist.  Initial examples involve hauling ore in open-pit mines, planting, maintaining crops and harvesting in farms, large warehouses, and transporting containers within ports. Trials, and even deployments, for additional specialized applications will continue in increasingly less controlled environments such as urban garbage collection.
  2. Trucks used in long-haul freight logistics. Several factors motivate the use of autonomous, and increasingly electrified, long-haul trucks for freight transportation between distribution centers that are outside city limits. These include: digital, global and highly optimized supply chains, shortage of long-haul truck drivers, fuel savings, driver safety, and supply chain economics. Utilizing freeways, and maybe even specially designated lanes in freeways, provides for environments that are more stable and consistent and less complex, compared to urban settings. The first significant tests for this case will start in 2018 and involve truck platooning.
  3. Vehicles used for short-haul package handling. The continued explosive growth of ecommerce is leading to the dramatic increase for package delivery and merchandise returns logistics. Logistics companies are looking for ways to effectively address last-mile on-demand delivery in urban and suburban environments. They are also exploring numerous different options  involving autonomous vehicles most, if not all, of which will be electrified. Tests of such vehicles have started and more are planned for the near future.
  4. Passenger shuttles. Autonomous, electrified passenger shuttles will come in different forms. Today they are providing transportation in a variety of university campuses, and other controlled environments that have low traffic, and impose low speed restrictions. Additional trials are on the way, including trials in retirement communities. Over time such shuttles will be introduced in increasingly complex urban settings to provide, first-mile/last-mile transportation to private and public mass transit systems, as well as transportation for the elderly and people with disabilities. Though the technology employed by the special-purpose autonomous vehicles is impressive, the business cases and associated models, as well as the necessary regulation have not yet been fully defined. Remember the Segway.  It employed an impressive array of technologies to address personal urban mobility but encountered several obstacles, including business model and regulation-related issues, in establishing a viable business case.
  5. Vehicles used for ride-hailing. Autonomous vehicles will provide numerous advantages to companies offering urban and medium-haul (up to 150 miles) ride-hailing and ride-sharing services. These include the ability to reduce operating costs, control the overall passenger experience, and reduce urban traffic congestion and pollution. While trials of growing scope will continue, the broad use of autonomous vehicles for such applications is several years away. These vehicles may not initially be electrified but ultimately they will evolve to be such.
  6. Vehicles used for private transportation. The growing proliferation of mobility services particularly those that blend on-demand with scheduled mobility (in the form of public and private mass transit) will negatively impact private car ownership but will not eliminate it. Automotive OEMs will definitely offer vehicles with Level 4 and Level 5 driving automation initially at their high-end models, and, as the price of the components that comprise the autonomous platform drops, to lower-price models that target broader segments of the market. Some of these vehicles may be shared.

These use cases lead me to two observations. First, two use cases (2-3) are about next-generation mobility for freight transportation. Three use cases (4-6) are about next-generation mobility for passenger transportation. Even as consumers continue to transition from vehicle ownership to vehicle access for transportation, autonomous vehicles will need to operate side-by-side with human-driven vehicles in streets and highways.

Second, five use cases (1-5) involve fleets rather than private vehicles. A case can be made on whether autonomous farming equipment will be privately owned or owned by fleet operating companies. The Fleet-Based On-Demand Shared Mobility value chain that I presented in previous posts can be employed in use cases 2-5. The new value chain will involve, impact, and potentially disrupt many industries beyond automotive. Industries such as energy, financial services, insurance, telco, utilities, and others are already being affected. Because of the previous observation, the new value chain will co-exist with the Vehicle Manufacturing and Sale, and Vehicle Use value chains that I presented in my book.

The “simpler” use cases will be deployed, and therefore monetized, before the more complex ones. The operators of specialized autonomous vehicles are already enjoying  operating cost reductions from their use. Once deployed, truck platooning will enable the reduction of fleet operating cost reduction due to fuel savings. Eventually, the biggest monetization opportunities will come from the use of autonomous short-haul and long-haul trucks, as well as from passenger vehicles employed for ride-hailing and ride-sharing services where I will focus for the rest of this post.

The Value Added By Big Data and AI

The fleet-based on-demand personal mobility value chain consists of four basic components as shown in Figure 1.Figure 1Figure 1: The value chain for fleet-based on-demand shared mobility[/caption]

While the central role of data and AI in making vehicle autonomy possible is beginning to be appreciated, their equally critical role in every aspect of fleet-based on-demand mobility services is less so. Big data and AI add value in every component of this value chain by automating many tasks, enabling insights-based decisions, and business models that are new to transportation. In order for some of these models to be possible, it will be necessary for extensive data sharing among the members of the new value chain that will need to be motivated by what I call in my book value exchange. An example of value exchange was recently introduced by Mitsubishi Motors. In order to collect driving data Mitsubishi has developed a mobile application for its car owners. In exchange for accessing the driving data collected by the application, the OEM will offer rewards such as oil changes and gift cards to the participating  drivers. It then monetizes the collected data by offering it to car insurance companies. Figure 2 below shows some of the potential data flows among the companies that will participate in the value chain.

Figure 2

Figure 2: The data flows across the fleet-based on-demand shared mobility value chain

Vehicle design, test, and manufacturing. Data from fleet creators, fleet operators, and fleet service and maintenance companies, can add significant value to the data about vehicle and component manufacturing. Coupling data-driven AI with 3D printing will lower the automakers’ manufacturing costs by optimizing material selection and component designs. It can also be used in order to improve vehicle designs, or to create environment-specific or function-specific designs, improve the testing and validation of the technology making possible vehicle autonomy, and their overall manufacturing process.

The application of predictive maintenance analytics to part failures will lower the manufacturer’s warranty costs, and will also be able to determine when changes due to wear and tear need to be made to a vehicle’s cabin. Knowing which parts fail and why allows for better designs and few recalls. It could also result in more effective marketing campaigns to fleet owners, fleet operators, as well as to consumers who will buy such vehicles for their private use.

Big data collected from test and production vehicles can be utilized by the machine learning systems that are incorporated in the Perception, Localization, and Planning software components of the Autonomous Vehicle Operation Platform to improve their performance. Such improvements lead to the transition from the “simpler” to the more complex of the six use cases described previously.

Simulation systems, such as those built by our portfolio company Metamoto, can fuse big data collected by fleet operators with simulated data and use AI. The fused data is used to test the performance of the predictive models incorporated in the various components of the Autonomous Vehicle Operating Platforms under a variety of “corner cases.” In this way these simulations help assess the performance of such platforms under conditions that may only be encountered infrequently but are important nonetheless in order to certify self-driving vehicles.

Fleet creation. Corporations that order vehicles from OEMs and then lease them to companies operating on-demand shared mobility fleets can realize important value by utilizing big data and AI. In particular, by analyzing

  • The data about the performance of each vehicle in specific fleets along with the various conditions (weather, road, etc.) under which this performance was observed (provided by fleet operators),
  • Vehicle configuration data for each vehicle in such fleets (provided by vehicle manufacturers and fleet operators), and
  • Prior financial and insurance data pertaining to the acquisition of the fleets (which they presumably have),

they can identify insights enabling them to negotiate lower per vehicle prices with manufacturers, and lower premiums with insurers on a region by region basis and on a fleet operator basis. They will also be able to determine how to continue to financially exploit a fleet’s vehicles once a lease with a fleet operator expires. In this way they will positively impact their overall fleet investments.

Fleet operation. The operators of fleets that are used for on-demand mobility services will differentiate themselves by the customer experience they will be able to offer, as well as the convenience, price, and safety of their service. Big data and AI are key enablers of all these, in the same way they are key enablers of autonomous mobility.

Next-generation mobility changes the Brand Loyalty model that has been used by incumbent automotive OEMs since the end of World War Two. Under the car ownership-centric model automakers measure brand loyalty by their ability to first protect and then increase their market share. They achieve this by retaining a consumer, or household, within their brands, and ideally, over time, transition them from the entry-level to their more expensive brands. This is the reason each automaker owns several vehicle brands. For example, in the case of Toyota this may mean moving a household that initially purchases, or leases, a Corolla to later leasing a Camry and eventually a Lexus. Under this model brand loyalty is measured every 3-10 years depending on how frequently the household changes vehicles, or acquires additional ones. During the intervening years the automotive OEM, at best, interacts infrequently with the household’s members. In fact, the OEM’s dealers interact more frequently with household members particularly during the manufacturer’s warranty period, building a richer relationship with them.

The transition from the car ownership-centric model to the hybrid model that combines car ownership with on-demand transportation services will change the prevailing notion of brand loyalty. Loyalty will not be measured by the willingness to continue purchasing a particular OEM’s vehicles, but will be measured by the miles traveled using an OEM’s vehicles and a fleet operator’s service. The new model will pit the OEM against the fleet operator for each consumer’s loyalty. The ongoing challenge for the OEM and fleet operators will be what share of the customer’s revenue-miles traveled they have at any point in time, which will translate to share of wallet. Big data and AI will add tremendous value in understanding loyalty, impacting customer experience during every interaction, but also in optimizing the service convenience, and increasing the revenue-miles traveled by each consumer. Customer experience extends outside the vehicle as well. Data-driven AI can be used for the creation of end-to-end personalized, multimodal ground transportation solutions. Our portfolio company Safegraph is developing such solutions.

A few examples of the big data that will be generated by fleets offering ride-hailing and ridesharing services using autonomous vehicles (including passenger shuttles) include:

  • Demand-based pricing for each trip,
  • Passenger profile data as well as information about the passenger(s) in each trip, e.g., was the passenger carrying luggage?,
  • Passenger wait times and travel times,
  • Vehicle idle miles vs revenue-generating miles,
  • In-cabin sensor data and vehicle sensor data, e.g., the cabin’s condition during each trip, health-related data from each passenger (alertness, body temperature, heart rate, perspiration, etc.),
  • Data collected from transportation and other infrastructures as the vehicle travels,
  • Weather and traffic conditions on each route taken.

By exploiting this data using AI fleet operators will be able to offer customer experience-enhancing capabilities such as personalize every vehicle’s cabin for each passenger in each trip, create multimodal transportation plans to fit each customer’s preferences and constraints, and optimize each transportation experience’s price to increase the share of wallet. The fleet operator can also use this data to optimize the overall fleet’s and each vehicle’s financial, and operational performance. This is done by maximizing each vehicle’s uptime and revenue-miles.

Fleet management, service, and maintenance. Performance data generated during each trip analyzed using AI technologies will improve each vehicle’s uptime while controlling its maintenance costs. Examples of the big data that can be collected and used by companies participating in this part of the value chain include:

  • Fuel consumption,
  • Tire condition,
  • Details about each trip the vehicle takes, including road conditions, traffic conditions, etc.
  • Each vehicle’s historical service, maintenance, and repair records,
  • Each vehicle’s historical accident and breakdown records.

By analyzing the combined fleet utilization and per vehicle fleet management, service, and maintenance data AI systems can optimize each vehicle’s maintenance and service in order to maximize its uptime at the lowest cost. In other words, such AI-based systems can predict how long to keep each vehicle in operation before bringing it in for service and maintenance. The optimization of maintenance results in better cost controls and better vehicle performance per location and even ride for both the fleet’s operator and the fleet’s maintainer. For example, because of the difference in the weather and road conditions, operating an autonomous vehicle in New York will be different than in San Francisco, or in Singapore. In addition, higher uptime implies that the vehicle is available over longer periods to generate revenue-miles. Airlines provide a very good example of what is possible by continuously analyzing such data and optimizing their fleet’s performance. Well-maintained vehicles also result in better customer experience that affects loyalty and further improves the fleet operator’ financial performance.

As a result of such analyses, fleet managers may be able to structure more economically advantageous contracts with fleet operators that will be the beneficiaries of higher vehicle uptime and better customer experience.

The Monetization Opportunities Of Big Data and AI

My firm invests in startups developing next-generation mobility solutions. In addition to the value that big data and AI provide to each company’s internal operations, I am interested in additional monetization opportunities for this data. Such monetization will result from the creation of new revenue streams. I have identified several such opportunities across this value chain.

Vehicle design, test, and manufacturing. Big data can provide two new revenue streams to vehicle OEMs. First, an OEM can offer to sell data they own (data ownership rights will be discussed in a later post) to companies in other parts of the value chain. For example, fleet creators and fleet operators could be interested in acquiring vehicle test data (real and simulated) in the process of determining the types of vehicles they should include in their fleets.

Second, because the Autonomous Vehicle Operating Platform and UX Platform must be constantly updated and extended to enable the vehicle to operate in new situations and environments, OEMs or the suppliers of such platforms can monetize such updates using transaction-, subscription-, or API-based business models. Similarly, using such models fleet operators and fleet managers will also be able to acquire from OEMs and other suppliers, e.g., HERE, new vehicle features. For example, using OTA updates in conjunction with subscription- and transaction-based business models Tesla is already monetizing the introduction of new features, such as battery range extenders, to its privately-owned vehicles.

Fleet creation. By analyzing each vehicle’s usage data, service records, and maintenance data, fleet leasing companies can determine the terms under which they will lease vehicles to a fleet operator in order to maximize their revenue opportunity while reducing their risk. For example, if the fleet owners detect that a fleet operator’s vehicles are frequently traversing hazardous areas, they may charge a premium to cover risks to the fleet operator. Conversely, if the fleet has a low accident rate, the owner may offer discounts in future leases.

In addition, fleet leasing companies can create new big data-related revenue streams by selling the data they collect to insurance and financial services companies that will be interested in using such data to enhance their own risk models, as well as to further expand their business activities.

Fleet operation. A variety of business models are being considered for the monetization of revenue-miles traveled using ride-hailing and ridesharing mobility services using autonomous vehicles. They include transaction-, subscription-, advertising-, and ecommerce-based models. The big data collected in the course or providing these services, combined with AI, offer several additional revenue-generation and cost-reduction opportunities to fleet operators.

The first new opportunity is from the use of data in Passenger CommercePassenger Commerce refers to the ecommerce performed by the passenger of autonomous vehicles while being transported. It can be conducted through the vehicle’s UX Platform or a passenger’s personal mobile device via:

  • Subscriptions to access content, e.g., an annual subscription to Apple Music, or a particular service, e.g., entertainment concierge.
  • Transactions for the purchase of goods, services, and content. For example, a traffic data provider, e.g., Inrix, can offer congestion data about a particular city on a particular day. An early example of such transaction-based commerce today is offered by Hertz for its Neverlost GPS and travel information platform. Hertz is able to charge extra for the vehicles that are equipped with this platform because of the information it provides to driver and passengers.
  • Advertising in exchange for free access to content, goods, and services. For example, rides to certain establishments, e.g., a specific restaurant or casino can be sponsored by the establishment itself.
  • Redemption of loyalty points. Automakers and fleet operators can reward their customers for their loyalty using a system similar to that used by airlines or hotel chains. These points can then be redeemed in much the same way these and other industries use such programs. For example, for every 5,000 ridesharing miles in a particular automaker’s vehicles, the consumer receives points that are redeemed towards a free meal from a particular food chain.

Big data is used to determine both what products and services a passenger will likely want to access and/or acquire, but also through what business model they would want to transact.

Fleet operators can also monetize the collected data using online advertising. Using data-driven approaches pioneered in the adtech industry, Uber already sells ads that are displayed on its mobile application. Fleet operators can also create additional revenue streams by licensing or selling parts of the big data they capture, as well as the insights they generate from this data.

Finally, in addition to monetizing this data by offering it to companies in this value chain, the data captured by fleet operators may also be monetized with companies that belong to other industries, e.g., airlines, utilities, financial services, and governments, just to list a few. For example, video data captured by the AV Platform’s cameras could be used by retailers to understand consumer traffic patterns in specific parts of a city.

Fleet management, service, and maintenance. Big data can provide two new revenue streams to fleet managers. First, using transaction-, subscription-, or API-based business models they can offer the data they collect about individual vehicles or even an entire fleet, and/or insights they derive from analyzing such data, to other companies in the value chain.  For example, vehicle OEMs would be interested in acquiring the maintenance data coming from the vehicles of their own brands and even competitor brands.

Second, because I expect that such companies may develop new big data/AI applications to automate their operations, they may consider licensing this software using subscription-based models. Smaller fleet management and maintenance would be interested in acquiring such software because they will likely not have the resources or expertise to develop it on their own.

The already established automotive business models continue to face margin compression and declining value. The models used today by companies offering ride-hailing and ride-sharing services are not offering better margins. By understanding the value of big data and AI in each component of the fleet-based on-demand shared mobility value chain as this will be utilized in the various autonomous vehicle use cases, we can identify several monetization opportunities that can result in higher-margin business models.

The next article in the series.

The previous article in the series.

3 thoughts on “Monetizing Big Data in Fleet-Based On-Demand Shared Mobility

Leave a Reply