DYNAMIC ADVERTISING CAMPAIGN BASED ON VEHICLE POSITION AND ROUTE DETERMINATIONS

Abstract

Computer-implemented procedure, including: Received, from a ride-sharing device (110; 302; 700; 800), a ride request that includes an origin and a destination; Determining a set of potential routes (254; 610) to fulfill the ride request; Analyzing the set of potential routes (254; 610) using a route target function that includes a weighted combination of a set of quality metrics to generate appropriate quality ratings for the potential routes (254; 610), wherein the route target function includes at least one passenger comfort parameter and at least one operational efficiency parameter; Processing at least a subset of the potential routes (254; 610) using an optimization process to improve at least a subset of the corresponding quality ratings; Determine, by means of a route determination system, a route (254; 610) in response to a ride request, wherein the route (254; 610) and a predicted time to the starting point (252) are based at least partially on route history data for a plurality of previously requested routes (254; 610) and the corresponding quality ratings, wherein each previously requested route (254; 610) is associated with a corresponding origin, destination and time period; Identifying one of the vehicles (334; 606) assigned to the route (254; 610) in response to the ride request; Determine, through the route determination system, a predicted arrival time for the assigned vehicle (334; 606) to reach the origin; Obtaining information for points of interest (POIs) within a distal area of ​​the origin; Determining promotional offers (614) for the POIs at least partially based on the predicted arrival time; and Transmitted to the passenger computer (110; 302; 700; 800) of computer-readable instructions relating to the promotions (614) for the POIs for display on the passenger computer (110; 302; 700; 800); and Update the weights of the quality metrics for the objective function based on the output of a machine learning model trained using route performance history data and current route performance data.

Description

[0001] The present invention relates to a computer-implemented method and a system for determining routes for passengers in vehicles and for promotional activities based on the determined route. GENERAL STATE OF THE ART

[0002] People are increasingly turning to services like ride-sharing to accomplish everyday tasks. Ride-sharing can involve assigning passengers to vehicles reserved for them for a specific period, or assigning them seats in vehicles that are also carrying other passengers. While individually assigned cars can offer some advantages, sharing vehicles can reduce costs and provide a degree of planning certainty. To ensure the profitability of such a service, it is often desirable to try to minimize costs and maximize vehicle utilization. Traditional approaches to determining which vehicle to assign to a particular trip or route consider the vehicles available at that time.However, such an approach may not be optimal, as the available vehicles could be located a considerable distance away, increasing the cost of providing that particular ride or route due to the additional expense of moving the vehicle to its origin. Furthermore, this extra distance can delay the ride's start time, which not only impacts the user experience but also reduces the vehicle's utilization, as it will be unavailable for a ride during the transfer time to the origin of the next route.

[0003] For example, US patent 2014 / 0173511A1 describes a vehicle pooling system in which potential passengers are offered a selection of rides. A processing unit considers information stored in memory regarding distances, travel times, and routes, and optimizes the latter using a stored algorithm. Furthermore, German patent DE 102009056641A1 describes a computer-aided method for route determination and ride-sharing, in which a server unit of a ride-sharing platform compares the pick-up and drop-off locations of potential passengers with possible freight routes and modifies potential routes, provided that the resulting route extensions remain below a predetermined value.Document US 2018 / 0053136A1 describes a computerized system that receives real-time transportation requests from user terminals and filters them using geographic location data to identify suitable transportation routes for rides. Document US 9715233B1 further describes a computerized taxi request system where various starting positions can be entered from a mobile device, and the taxi route is then determined based on these positions, taking into account any road restrictions. Document US 9726510B2 further describes a navigation system that uses a mobile computing unit's calendar to define a starting point and a destination, calculates a route, and then analyzes potentially relevant intermediate stops along the way. Based on identified potentially relevant intermediate stops, several alternative routes are offered.

[0004] The present invention is based on the objective of creating improved methods and systems of the aforementioned type, avoiding the disadvantages of the prior art and admirably developing the latter further. In particular, an improvement is to be achieved with regard to the problems described, namely, on the one hand, optimizing travel and ride-sharing costs and, on the other hand, avoiding a reduction in vehicle utilization.

[0005] According to the invention, the aforementioned problem is solved by a computer-implemented method according to claim 1, a computer-implemented method according to claim 9, and a system according to claim 14. Preferred embodiments of the invention are the subject of the dependent claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Various embodiments according to the present disclosure are described with reference to the drawings, in which the following applies: Fig. Figure 1 illustrates an exemplary ride-hailing environment in which different embodiments may exist. The Fig. 2A and Fig. 2B illustrates exemplary origin and destination locations and routes for serving these locations, which can be determined according to different embodiments for a service area over a period of time. Fig. Figure 3 illustrates an exemplary system that can be used to implement aspects of the different embodiments. Fig. Figure 4 illustrates an exemplary process for determining dynamic promotions for an optimized route solution for a set of journey requirements, which can be used according to different embodiments. Fig. Figure 5 illustrates an exemplary process for determining dynamic promotions during a predicted period for proposed route solutions, which can be used according to various embodiments. Fig. Figure 6 illustrates an exemplary process for determining dynamic promotions when a delay is detected for proposed route solutions, which can be used according to various embodiments. Fig. Figure 7 illustrates an exemplary computing device that can be used according to various embodiments to transmit driving requests and receive route options. Fig. Figure 8 illustrates exemplary components of a computing device that can be used to implement aspects of the different embodiments. DETAILED DESCRIPTION

[0007] The following description details various embodiments. For illustrative purposes, specific configurations and details are presented to provide a thorough understanding of these embodiments. However, it will also be apparent to those skilled in the art that the embodiments can be implemented without these specific details. Furthermore, well-known features may be omitted or simplified to avoid obscuring the described embodiment.

[0008] The approaches described and suggested herein concern the provision of transportation in response to various demands. In particular, different embodiments provide approaches for determining and selecting from various promotions and route solutions to serve a set of transportation demands. The demands may involve the transportation of people, animals, packages, or other objects or passengers from a point of origin to a destination. The demands may also include at least one time component. A provider, such as a transportation service, may use an objective function to balance various metrics when selecting between proposed route solutions to serve a set of customer trip demands.An objective function can provide a compromise between, for example, the passenger experience and the provider's profitability, taking into account metrics such as passenger comfort, operational efficiency, and the ability to serve confirmed trips. The analysis can include not only planned trips or trips currently being planned, but also trips that are currently in progress. One or more optimization processes can be applied, which can vary the component values ​​or weights of the objective function to attempt to improve the quality score generated for each proposed route solution. A solution can be selected for implementation, at least in part, based on the resulting quality scores of the proposed route solutions.

[0009] In addition to optimizing routes for various requirements, approaches of different implementations can also implement proactive solutions to compensate for predicted delays or waiting times in response to a specific trip request. For example, there may be an anticipated delay, intentional change, or necessary adjustment that negatively impacts the service quality of a particular trip. This could include, for example, making an additional stop, obtaining a different vehicle, or delays due to a driver change. In some cases, it may also involve relocating a passenger to a different vehicle or route due to overbooking or other similar issues.Approaches based on various embodiments can proactively determine reasonable compensation for the disruption to the service that is likely to be acceptable to the affected driver(s). In some embodiments, the types of compensation may differ between passengers. This provides alternatives to incurring unreasonable costs or causing inconvenience to multiple passengers when the service provider or operator may have to cancel the trip or deliver it with a significant delay, but proactively offers compensation to account for the change. There may also be unexpected occurrences, such as an accident, a detour, or other event, that could cause an unexpected delay to the previously determined route and arrival time.In response to a specific ride request being delayed, the provider can dynamically determine a promotional solution, such as a voucher for a coffee at a nearby café, to compensate for the delay. The level of compensation for an unexpected delay or a delay due to a factor other than the provider may be less than the compensation provided for an anticipated or intentional disruption to the service. As discussed herein, the appropriate type and level of compensation can then be determined not only for the specific passenger receiving compensation but also for the nature of the disruption to the service or the inconvenience experienced.In at least some embodiments, the ability to offer compensation for offer adjustments expands the scope of the route optimization solution, as the available options can become more flexible, at least if passengers are open to such compensation options. Where available, a dynamic promotion can be provided using offers from providers in close proximity to the requester, allowing the provider to reduce overall inefficiencies of the ride-sharing and route system, improve its optimization processes based on data and preferences from its user base gathered through responses to the provided promotions, and increase productivity. Promotions in such an approach can be targeted at both supply and demand.

[0010] Various other such functions may also be used within the scope of the various embodiments, as would be obvious to a person skilled in the art in view of the teachings and suggestions contained herein.

[0011] Fig. Figure 1 illustrates an exemplary environment 100 in which aspects of the various embodiments can be implemented. In this example, a user can request transportation from an origin to a destination, for example, using an application running on a client computing device 110. Various other approaches to transmitting requests, such as messaging or telephone mechanisms, can also be used within the scope of the different embodiments. Furthermore, at least some of the requests can be received by or on behalf of an object that is being transported or whose transport is scheduled.For example, a client device can be used to transmit an initial request for an object, package, or other deliverable, and subsequent requests could then be received, for example, from the object or from a device or mechanism associated with the client device. Other forms of communication can be used instead of requests, which may include instructions, calls, commands, and other data transmissions. For various embodiments discussed herein, a "client device" should not be understood in the strict sense as a conventional computing device unless otherwise specified, and according to various embodiments, any device or component capable of receiving, transmitting, or processing data and communications may function as a client device.

[0012] Transportation can be provided using a vehicle 100 (or other object) capable of transporting one or more passengers simultaneously. Although, as used herein, passengers often refer to human occupants, it is understood that in various embodiments a "passenger" can also refer to a non-human passenger, which may include an animal or an inanimate object, such as a package for delivery. In this example, a ride-sharing service offers routes that use at least one type of vehicle that includes space for a driver 102 and seats or other capacity for up to a maximum number of passengers.It is understood that various types of vehicles with different capacities or configurations can be used, and that autonomous vehicles without dedicated drivers can also be used within the scope of the various embodiments. Vehicles such as smart bikes or personal transportation vehicles, which may have seating capacity for only one passenger or a limited number of passengers, can also be used. For a given vehicle on a given route, a number of available seats (or other passenger positions) may be occupied by passengers, while another number of seats may be unoccupied. In some embodiments, objects such as packages or delivery items may also occupy free space for a journey.To improve the profitability of ride-sharing services, it may be desirable for occupancy in at least some configurations to be as close to full as possible throughout the entire journey. Such a situation results in very few unsold seats, thereby improving operational efficiency. One way to achieve high occupancy could be to offer only fixed routes where all passengers board at a fixed origin and disembark at a fixed destination, without passengers boarding or disembarking at intermediate locations.

[0013] In the present example, a given user can enter a point of origin 112 and a destination 114, among other such options, either manually or from a set of suggested locations 116, such as by selecting on a map 118 or other interface element. In further embodiments, a source, such as a machine learning algorithm or an artificial intelligence system, can select the appropriate locations based on relevant information, such as a history of user activity, the current location, and the like. Such a system can be trained using historical ride-sharing data and can learn and improve over time by using more recent ride-sharing and passenger data, among other such options.A backend system or other service provided by a service provider can access this information and attempt to match the request with a specific vehicle that has capacity at the appropriate time. As is known for such purposes, it may be desirable to select a vehicle that will be close to the starting point at that time to minimize overhead costs, such as fuel and driver expenses. As mentioned, capacity, among other such metrics, can include a seat for a passenger or sufficient available volume for a package or object to be transported.

[0014] However, such an approach may not be optimal in all situations, as it can be difficult to persuade enough users or object providers to agree to be at a specific origin at a specific time or within a specific time window, potentially resulting in relatively low occupancy or capacity utilization and thus low operational efficiency. Furthermore, such an approach may result in fewer rides being provided, which can reduce overall revenue. Requiring multiple users to travel to a specific, fixed origin may also lead those users to use other modes of transportation, including taxis or dedicated ride-sharing vehicles, which do not require the additional effort. Accordingly, in at least some embodiments, it may be desirable to incorporate rider convenience into the selection of routes to be provided.What is convenient for one passenger may be inconvenient for others. For example, picking up a passenger from their home could add an extra stop and a longer route to an existing one, which might be unacceptable to passengers already on or assigned to that route. Furthermore, different passengers may prefer to depart from different locations at different times and reach their destinations within a specific timeframe, meaning that the interests of these passengers are, at least to some extent, in competition with each other and with those of the ride provider.Therefore, in at least some embodiments, it may be desirable to balance the relative experience of different passengers with the economics of the ride-sharing service for specific trips, routes, or other transportation options. In some embodiments, the provider may, based on passenger preferences, suggest that a particular passenger be picked up at an alternative location and offer a promotion as an incentive for the passenger to accept the alternative pickup location, thereby increasing the efficiency of the overall route. The provider may also provide dynamic promotions to a passenger to compensate for anticipated unexpected delays in response to their ride request.While such an approach will likely prevent a ride provider from maximizing profit per ride, there can be a middle ground that allows the service to be profitable while providing at least a satisfactory service to the various passengers or users. Such an approach can improve the passenger experience and lead to higher passenger numbers, which, if managed properly, can increase revenue and profit.

[0015] The Fig. 2A and Fig. Figure 2B illustrates an exemplary approach that can be used to provide such a service according to various embodiments. In the exemplary map 200 from Fig. 2A specifies a set of origin points 202 and destination points 204 over a given period of time, indicating locations between which one or more users would like to travel. As illustrated, there are clusters of locations where users might want to be dropped off or to which items need to be delivered, and these can correspond to city centers, urban areas, or other regions containing a number of different businesses or other destinations. Origin locations, however, can be less clustered, as might be the case with suburbs or rural areas where passengers' homes may be located. Furthermore, clustering can vary throughout the day, such as when people travel from their homes to their workplaces in the morning and generally in the opposite direction in the evening.Between these time periods, clustering may be minimal, or it may primarily involve locations within an urban area. Particularly in urban areas, there may be a number of vendors, shops, retailers, and restaurants in close proximity to origin points. For example, passengers requesting rides after work may have origin points at their offices, which often have cafes or shops nearby. The provider can identify and log these vendors to offer passengers dynamic promotions to compensate for route delays and / or as an incentive for passengers to adjust their origin point to achieve an optimal overall route.

[0016] In the example route map from Fig. 2B For a starting point 252, there can be a multitude of routes 254, determined based on the destination and / or other users and their destinations and origins identified as being assigned to a particular vehicle. To determine the routes to be provided, as well as the vehicles (or types of vehicles) to be used to provide these routes, various factors can be considered, as discussed and suggested in this document. A function of these factors can then be optimized to provide an improved customer experience or an improved transportation experience for transported objects, while also providing improved profitability or at least improved operational efficiency with respect to other available routing options.The optimization approaches and route suggestions can be updated over time based on other available data, which may include more recent trip data, ride-sharing requests, traffic patterns, roadwork updates, and the like. In some embodiments, an artificial intelligence-based approach, which may include, for example, machine learning or a trained neural network, can be used to further optimize the function based on various trends and relationships determined from the data, as discussed elsewhere herein.

[0017] Approaches according to various embodiments can utilize at least one objective function to determine route options for a set of vehicles or other transportation mechanisms for one or more service or coverage regions. At least one optimization algorithm can be applied to adjust the various factors considered in order to improve an outcome of the objective function, such as minimizing or maximizing the rating for a set of route options. Optimization can, for example, apply not only to specific routes and vehicles but also to compensating for unexpected delays by providing promotional offers to enhance the passenger experience and / or incentivizing passengers to adjust their point of origin due to delays on optimized routes.An objective function can serve as an overall measure of the quality of a route solution, a set of suggested route options, or past route choices. It codifies a desire to balance various significant factors, which, among other options, may include passenger comfort or experience, service delivery efficiency for a given area, and adherence to quality of service (QoS) for specific trips. For a given set of origins and destinations over a given period, the objective function can be applied, assigning a score to each suggested route solution, such as an optimized route score, which can then be used to select the optimal route solution.In some embodiments, the route option with the highest route rating is selected, whereas in other embodiments, approaches may be taken to maximize or minimize the resulting rating, or to generate a ranking along with various other rating, ranking, or selection criteria. Route options with the lowest rating may also be selected, such as when the optimization function is optimized based on a cost measure that should be as low as possible versus a factor, such as a profit measure that should be as high as possible, alongside other such options. In other embodiments, the selected option may not have the highest optimal objective rating, but rather an acceptable objective rating while fulfilling one or more other trip selection criteria, which may relate to operational efficiency or minimum passenger experience, among other things.In one embodiment, an objective function takes passenger comfort, the ability to complete confirmed trips, fleet operational efficiency, and current demand as inputs. In some embodiments, weights are provided for each of these terms, which are learned over time, for example, through machine learning. The factors or data comprising each of these terms or values ​​can also change or be updated over time.

[0018] For example, a passenger comfort rating can take various factors into account. One factor could be the distance from the passenger's requested origin point to the origin point of the selected route. The rating can be performed using any relevant approach, where, for example, an exact match receives a rating of 1.0 and any distance greater than a maximum or predefined distance receives a rating of 0.0. The maximum distance could be, among other options, the maximum distance a user is willing to walk or drive to an origin point, or the average maximum distance of all users. For packages, this could be the distance a provider is willing to travel to have those packages delivered to their respective destinations.The relationship between these factors can also vary; for example, a linear or exponential function can be used. For instance, in some implementations, an origin located halfway between the requested and proposed origins might be assigned a comfort rating of 0.5, whereas in other approaches, it would only reach 0.3 or less. A similar approach can be used for time, where the time interval between the requested and proposed pickups can be inversely proportional to the applied comfort rating. Various other factors can also be considered, including trip length, number of stops, target time, anticipated traffic, and other such factors. The comfort score itself can be a weighted combination of these and other such factors.Depending on the passenger comfort score, which is balanced against the overall route rating, the provider can dynamically determine a promotion to incentivize the passenger to update their origin. For example, if the comfort score is within a certain threshold and closer to 0.0 (i.e., least comfortable for the user), the likelihood of the passenger being receptive to an updated origin or accepting the ride decreases. To combat the risk of the passenger canceling or having a negative experience, the provider can offer a promotion, such as a free coffee at a café along the way to the updated origin, to incentivize the passenger.

[0019] Optimizing, or at least considering, a passenger's comfort metric can help ensure that rides offered to passengers are at least competitively comfortable. While passenger comfort can be subjective, the metric can consider objective factors to determine whether the comfort is competitive compared to other available transportation options. Any suitable factors that can be objectively determined or calculated using available data can be included. These factors might include, for example, the ability (or inability) to provide different ride options. The factors might also include a difference in departure or arrival time based on the time(s) requested by the passenger for the route.In some implementations, a fellow passenger can provide a target time, while in others, passengers can provide time windows or acceptable ranges, among other such options. Another factor can concern relative travel delay, either as expected or based on historical data for similar routes. For example, certain routes through certain high-traffic areas may have variable arrival times, which can be factored into the comfort rating for a potential route through that area or those areas. Another factor can concern the walking (or off-route) required of the user for a given route. This can include, as mentioned, the distance between the requested origin and the suggested origin, as well as the distance between the requested destination and the suggested destination.Any walking distance required for transfers between vehicles can also be taken into account. The provider can identify and offer promotions to compensate for delays, such as a free or discounted magazine or book at a nearby store. In another example, the provider can offer a menu of promotions for various vendors along the planned walking route. The passenger can then select the most attractive or relevant promotion at that time, thus enhancing the passenger experience and satisfaction.

[0020] Another factor that can be considered in passenger comfort metrics, but which may be more difficult to measure, is the desirability of a particular location. In some implementations, the rating can be determined by a service provider employee, while in others, a rating can be determined based on critiques or feedback from various fellow passengers, among other options. Several factors can be considered when assessing the desirability of a location, such as the terrain type or traffic associated with that location. For example, a flat location might receive a higher rating than a location on a steep mountain.Furthermore, the availability, proximity, and type of smart infrastructure can also influence the rating, as locations near or managed by smart infrastructure may be rated higher than those without such proximity, since these areas can provide more efficient and environmentally friendly transportation options, among other such benefits. Similarly, a location with low pedestrian traffic might receive a higher rating than one near a busy intersection or tram tracks. In some implementations, a safety metric may be considered, determined based on data such as crime statistics, visibility, lighting, and customer feedback, among other options.Various other factors can also be taken into account, such as the proximity of train lines, retail stores, cafes, and the like, for which the provider can determine and offer promotional offers to the passenger. These promotions aim to encourage the passenger to accept an alternative location, which may be less convenient for the user but is ultimately more desirable for the provider in terms of maintaining the overall route rating and efficiency.

[0021] As mentioned, in some embodiments a route optimization system may attempt to use such an objective function to determine and compare different route options. Fig. Figure 3 illustrates an exemplary system 300 that can be used to determine and manage vehicle routes according to various embodiments. In this system, different users can use applications running on different types of computing devices 302 to transmit route requests over at least one network 304 so that they can be received by an interface layer 306 of a service provider environment 308. The computing devices can be any suitable devices known or used for transmitting electronic requests, including, for example, desktop computers, notebook computers, smartphones, tablet computers, and wearable computers, among other such options.The network(s) may include any suitable network for transmitting the request, such as any selection or combination of public and private networks using wired or wireless connections, such as the Internet, a cellular data connection, a WLAN connection, a local area network (LAN), and the like. The service provider environment may include any resources known or used for receiving and processing electronic requests, such as various computer servers, data servers, and network infrastructures discussed elsewhere herein.The interface layer can include interfaces (such as application programming interfaces), routers, load balancers, and other components useful for receiving and forwarding requests or other communications received in the service provider environment. The interfaces and the content displayed through these interfaces can be provided using one or more content servers capable of serving content (such as web pages or map tiles) stored in a content data store or other similar repository.

[0022] Request information can be directed to a route manager 314, which may contain code running on one or more computing resources. This route manager is configured to manage aspects of routes to be provided using various vehicles from a vehicle pool or fleet associated with the transportation service. The route manager can analyze request information, determine available planned routes from a route memory 316 that has a certain capacity, match the request criteria, and provide one or more options back to the appropriate device 302 for selection by the potential passenger. The suggested suitable routes can be based on various factors, such as proximity to the origin and destination of the request, availability within a specific time window, and the like.In some embodiments, an application on a client device 302 may instead present the available options from which a user can choose, and may instead include the request to obtain a seat for a specific planned route at a specific planned time.

[0023] As mentioned, in some embodiments, users can either suggest route information or provide information corresponding to a route desired by the user. This might include, for example, a point of origin, a destination, a desired pick-up time, and a desired drop-off time. Other values ​​can also be provided, such as a maximum duration or trip length, a maximum number of stops, permissible deviations, and the like. In some embodiments, at least some of these values ​​may have maximum or minimum values ​​or permissible ranges specified by one or more route criteria. Furthermore, there may be various rules or regulations dictating the extent to which these values ​​may change under different circumstances or situations, such as for specific types of users or locations.The route manager 314 can receive multiple such requests and can attempt to determine the best selection of routes to meet the various requirements. In this example, the route manager 314 can work with a rating generator 318, which calculates a route rating for each route identified by the route manager 318 as a feasible option in response to the ride requests. The rating generator 318 can calculate ratings, taking into account options with varying numbers of vehicles, different vehicle selections or placements, and different options for how the various customers are transported to or near their desired destinations at the desired times.It is understood that in some embodiments, customers may also request specific locations and times where no deviation is permitted, and the route manager may either have to determine an acceptable route option or reject the request if a minimum set of criteria is not met. In some embodiments, an option may be provided for each request, and a pricing manager 322 may determine the cost for a specific request using pricing data and policies from a pricing archive 324, which the user may then accept or reject.

[0024] In this example, the rating generation module 318 can calculate route ratings based on an objective function. In various embodiments, the objective function is applied to each potential route to generate a "route quality rating" or other such value. The values ​​of the various options can then be analyzed to determine the routing options to be selected. In one embodiment, the route optimization module 320 applies the objective function to determine the route quality ratings and can then select the set of options that provides the highest overall quality rating or the highest average overall quality rating. Various other approaches can also be used, as the average person skilled in the art will understand in light of the teachings and suggestions contained in this document.The rating generation module 318 can communicate with a route optimization module 320, which can perform an optimization process using the provided route options to determine an appropriate set of routes to be provided in response to the various requirements. In some embodiments, the route optimization module 320 can determine a more optimal route from an alternative origin to the destination, where the alternative route has a better route rating calculated by the rating generation module 318.

[0025] Optimization algorithms can evolve automatically over time, for example, through random trials or based on various heuristics. As these algorithms evolve, the value of the objective function can serve as a measure of the suitability or value of a new generation of algorithms. Algorithms can change over time as service areas and riders require adjustments, and to improve under the same or similar conditions. Such an approach can also be used to anticipate future changes and their impact on the service, as well as how various factors will change. This can help determine the need to add more vehicles, relocate parking locations, and so on.In some implementations, approaches incorporating artificial intelligence, such as those using machine learning, can be used with optimization algorithms to further improve performance over time. For example, increasing and decreasing various factors, such as ridership levels, customer ratings, and the like, as well as actual costs and times, can lead to changes that can be fed back into a machine learning algorithm to determine the appropriate weights, values, ranges, or factors to use with an optimization function.In some embodiments, the optimization function itself can be generated by a machine learning process that considers the various factors and historical information to generate a suitable function and further develops this function over time based on more recent result and feedback data as the machine learning model is further trained and able to develop and recognize new relationships.

[0026] The service provider environment 308 can also include a promotion engine 310 that accesses a vendor database 312, in which vendors can be selected based on the origin in the request. The promotions can be generated based on the routes selected by the route manager 314 and, in some embodiments, can incorporate the route ratings from the rating generation module 318. For example, the set of routes generated by the route manager 314 for a specific request that has an origin and a destination may not have a route rating above a minimum threshold. Alternatively, the route optimization 320 can determine more optimal routes from an alternative origin with higher route ratings.To incentivize the customer to accept the alternative origin in order to meet a route evaluation threshold for a more optimized route, the promotion engine 310 can generate promotions for salespeople identified within the immediate vicinity of or en route to the alternative origin. In another embodiment, a fleet manager 330 can track the fleet of vehicles and / or a specific vehicle 334 assigned to the route 332 and determine that there is a delay on the route. To compensate for the delay in arriving at the customer's origin, the promotion engine 310 can provide a promotion for a salesperson located near the user's origin.The promotion engine 310 can communicate with an account manager 326 with access to a user database 328 to personalize the promotions provided to the customer; for example, the promotions can be personalized based on personal preferences, customer history data, age, gender, geographic location and / or other demographic data associated with the customer.

[0027] As discussed elsewhere in this document, various types of promotions or compensation may be available from which a promotion engine or other such system or service may select. Each option may have an associated monetary value, desirability factor, or other measure of value that can be used to determine appropriate compensation for a given situation. Likewise, factors or aspects of the data may be present that can be used to select promotions or compensation for different passengers or entities, which, among other such options, may be based on sales or preference data. In some embodiments, the promotions may be targeted at the mobility provider's offering, while in other situations, the promotions may be targeted at the passengers of the mobility provider's service.For example, promotions aimed at riders can be directly related to mobility, such as offering a free ride to compensate for an unacceptable or poor experience on a completed trip. Ride-focused promotions can also be indirectly related to mobility, such as allowing a rider to be driven closer to a specific café, regardless of the impact on the charged service quality for the ride. Other promotions may also be provided, such as traditional location-based in-app or in-vehicle marketing, or location-independent marketing. Various other types of promotions may also be provided, such as those that attempt to entertain the user or allow the user to "kill time" during the ride or while waiting for a pickup or connection.These might include, for example, games, puzzles, videos, or podcasts available through the provider's app or another similar source. Other offerings may be less directly promotional but could ask questions about the user's interests or whether they have any impressions of certain local businesses, which can entertain the user while allowing future promotions or offers to be of greater interest to the passenger. Various other promotions and types of compensation may also be offered, as would be obvious to the average professional given the disclosure at hand.

[0028] Fig. Figure 4 illustrates an exemplary process 400 for determining dynamic promotions for an optimized route solution for a set of trip requests, which can be used according to various embodiments. In this exemplary process 400, the route optimization system can receive trip requests from the customers' computing devices. The trip request can correspond to a future period for a geographic service area and include at least one origin and one destination. In some embodiments, the trip request can also include other parameters, such as customer preferences or time constraints. In response to the trip requests, available vehicles can be determined 404, and a set of route solutions can be determined based on the available vehicles and the information from the trip requests 406.For each route in the set of route solutions, a route score can be calculated. 408 Based on the route scores of the routes in the route solutions, a solution with an alternative origin can be identified as having a better route score. 410 To incentivize the customer to travel to the alternative origin, the route system can provide a promotion. To provide a relevant promotion, map information can be analyzed to identify vendors within a certain proximity of the alternative origin. The analysis can also consider vendors' opening hours, the goods and services they provide, and their relevance to the customer. Promotions for the identified vendors can be generated and provided to the customer. 414As described elsewhere herein, promotions can also be determined based on customer preferences, route history, or promotions accepted by the customer. Additionally, the promotion can be determined based on the estimated time it takes the customer to complete the promotion transaction and reach the alternative location. For example, the promotion might be for a salesperson en route to an alternative origin location, and considering the time the customer needs to complete the transaction can still result in a higher route score, as the assigned vehicle can now pick up the customer at the alternative origin location.

[0029] Promotions for various specific sellers can be made available to the customer for selection. 414. The promotions can be displayed on the customer's computing device for selection, for example, as a menu of promotions displayed on the customer's mobile phone. The promotions may include an offer for a discount or a free item from a specific seller and may provide the location so that the customer can decide whether or not to accept the promotion. The menu of promotions may give the customer the option to search for promotions with other sellers near the point of origin or an alternative point of origin, provided that the route optimization system has determined that the corresponding route score is still above a minimum threshold so as not to compromise the quality of the service provided to the customer.If the customer does not accept any of the promotions 416, the route optimization system may return to determine another set of route solutions with better route ratings 406, identify other alternative origin locations 410, and / or determine other vendors to provide promotions for them 412. Alternatively, the customer may retain the original route with the origin location, and the assigned vehicle will pick up the customer at the original origin location 420. If the customer accepts one of the promotions 416, then the route with the alternative origin location will be selected 418 so that the assigned vehicle can proceed with it.

[0030] Fig. Figure 5 illustrates an exemplary process 500 for determining dynamic promotions during a predicted period for proposed route solutions, which can be used according to various embodiments. In another embodiment, the route system can receive a trip request 502 that includes at least one origin and one destination. The trip request can include other parameters set by the requester, such as a required vehicle type or number of passengers. History route data can be analyzed 504 to determine at least one route in response to the trip request to transport the requester from the origin to the destination. As discussed elsewhere in this document, the determined route can be based on a route evaluation that includes an algorithm that weights various factors, including, among others, previous route information (e.g.,The route optimization system considers factors such as average time to complete the route, customer preferences, origin, destination, other customers, traffic conditions (e.g., including traffic history and real-time tracking), customer comfort, fleet availability, etc. It can identify a set of potential routes and select the one with the best route rating, or select routes with a route rating above a minimum threshold.

[0031] An assigned vehicle can be identified for the specified route 506. In some embodiments, a vehicle can first be identified before the route is determined; for example, the vehicle can be selected based on its proximity to the origin in the trip request. Subsequently, a predicted time for the assigned vehicle to arrive at the origin is determined 508. Additionally, the vehicle can be identified based on driver ratings, the vehicle's passenger and / or cargo capacity, etc. Map information can be analyzed 510 to determine vendors within a specified proximity of the origin, taking into account the predicted timeframe for the assigned vehicle to arrive at the origin. For example, within 100 feet of the origin or within a 3-minute walk, depending on the predicted time of arrival of the vehicle.The system can then generate and deliver promotions to customers for sales associates located at the origin location 512. These promotions can be personalized based on customer preferences, historical data, the predicted time, and / or geographic location. For example, if the predicted time is only two minutes, the promotion might be for a coffee or another quick transaction. However, if the predicted time is ten minutes, the promotion might be for a magazine the customer can read while waiting or a coupon for a clothing store to check while waiting for their assigned vehicle. The promotions can be transmitted to the customer's device for viewing and selection, allowing the customer to scroll through a menu of promotions and corresponding sales associates.

[0032] Fig. Figure 6 illustrates an exemplary process 600 for determining dynamic promotions when a delay is detected for proposed route solutions, which can be used according to various embodiments. In this exemplary process, one or more customers can submit a ride request to the route optimization system 602. Each ride request can include at least one origin and one destination. The ride request can additionally include other parameters and / or information specified by the customer, for example, geographic area, number of passengers, customer demographic information, etc. A route can then be determined in response to the ride request 604, which can be based at least partially on customer history information, such as previous ride requests, preferences, and / or settings.The identified route can be selected from a set of routes determined by the route optimization system based on trajectory routes from this origin or similar locations, trajectory routes to the destination or similar locations, trajectory routes traveled by the customer, trajectory traffic conditions, current traffic conditions, and so on. As discussed elsewhere in this document, the identified route can be determined based on a route score, for example, the highest score from the set of routes or a score above a minimum threshold. The route score can be calculated using an algorithm that weights various factors, including, but not limited to, service quality, trajectory routes, trajectory times to complete routes, customer comfort, the desirability of the origin and / or destination, fleet availability, and so on.

[0033] An assigned vehicle can then be identified to complete the specified route in response to the trip request. The assigned vehicle can be identified based on at least the driver rating, the driver's location, fleet availability, proximity to the origin, current vehicle capacity, and / or the specified route. Alternatively, in another embodiment, the assigned vehicle can be identified prior to determining the route. For example, a vehicle can be identified as traveling toward the origin of the trip request and thus be identified as the assigned vehicle. Depending on the assigned vehicle's trajectory, the route can be determined to minimize the number of turns or backtracking required by the assigned vehicle, thereby improving overall efficiency, productivity, and customer comfort / experience.Once the route and assigned vehicle are determined, the vehicle's predicted time of arrival at the point of origin can be calculated (608). The predicted time can be provided to the customer. The route optimization system can track the assigned vehicle along the route (610), which can also be displayed to the customer on their device. Upon detecting a change in the predicted time of arrival (612), the route optimization system can generate and provide promotional offers to sales representatives located at the point of origin (614). A change in the predicted time of arrival may include unexpected delays due to driver error, traffic conditions, detours, weather conditions, etc.Depending on the cause of the delay, the route optimization system can use the data to update its optimization and selection of routes, as well as its calculation to determine the predicted times.

[0034] Promotional offers can be made available to the customer for display on their device, for example, as a selectable menu. These offers might include a discount or a free item from a specific vendor and can provide the vendor's location, allowing the customer to decide whether or not to accept the promotion. To deliver a relevant promotion, map information can be analyzed to identify vendors within a certain radius of the customer's location. The analysis can also consider vendors' opening hours, the goods and services they offer, and their relevance to the customer.The Promotions menu may give the customer the option to search for promotions with other vendors near the point of origin or an alternative point of origin, provided that the route optimization system has determined that the promotion transaction will not cause further delay but will compensate for the change in the predicted time of arrival. If the customer does not accept any of the promotion(s) 616, the route optimization system may return to determining an alternative route 604 or an assigned vehicle 606 and / or generating promotions 614. Alternatively, the route optimization system may determine an alternative point of origin with alternative promotions for alternative vendors 620.If the customer accepts one of the promotions 616, the assigned vehicle can proceed on the specified route 618, with the updated predicted arrival time coinciding with the time the customer needs to complete the selected promotion.

[0035] Fig. Figure 7 illustrates an exemplary computing device 700, which can be used according to various embodiments. Although a portable computing device (e.g., a smartphone or tablet computer) is shown, it is understood that, according to various embodiments discussed in this document, any device capable of receiving, processing, and / or transmitting electronic data may be used. Such devices may include, for example, desktop computers, notebook computers, smart devices, Internet of Things (IoT) devices, video game consoles or controllers, wearable computers (e.g., smartwatches, glasses, or contacts), set-top boxes for televisions, and portable media players.In this example, the computing device 700 has an outer casing 702 that covers the various internal components and a display screen 704, such as a touchscreen, capable of receiving user input during operation of the device. These can also be additional display or output components, and not all computing devices include display screens, as is known in the field. The device can include one or more network or communication components 706, such as at least one communication subsystem to support technologies like cellular communication, WLAN communication, or Bluetooth. ® This may include communication and so on. Wired connections or links for connecting via a landline or other physical network or communication component may also be present.

[0036] Fig. Figure 8 illustrates an exemplary set of components comprising a computing device 800, such as the one relating to Fig.The device described in Section 12 may include computing devices for other purposes, such as application servers and data servers. The illustrated exemplary device includes at least one main processor 802 for executing instructions stored in physical memory 804 on the device, such as, among other such options, dynamic random-access memory (DRAM) or flash memory. As would be apparent to the average person skilled in the art, the device may also include many types of main memory, data storage, or computer-readable media, such as a hard disk drive or solid-state storage, which acts as data storage 806 for the device. Application instructions for execution by the at least one processor 802 may be stored by data storage 806 and then loaded into main memory 804 as necessary for the operation of the device 800.In some embodiments, the processor may also include internal memory for temporarily storing data and instructions for processing. The device may also support removable storage, which is useful for sharing information with other devices. The device also includes one or more power components 810 for supplying power to the device. These power components may include, for example, a battery compartment for supplying power to the device using a rechargeable battery, an internal power supply, or a connector for receiving external power, among other such options.

[0037] The computing device may include or communicate with at least one type of display element 808, such as a touchscreen, an organic light-emitting diode (OLED), or a liquid crystal display (LCD). Some devices may include multiple display elements, which may also include, for example, LEDs, projectors, and the like. The device may include at least one communication or networking component 812, which may enable, for example, the transmission and reception of various types of data or other electronic communications. Communication may take place over any appropriate type of network, such as the Internet, an intranet, a local area network (LAN), a 5G or other cellular network, or a wireless network, or may use transmission protocols, such as, but not limited to, Bluetooth. ®or NFC. The device may include at least one additional input device 814 capable of receiving input from a user or other source. This input device may include, for example, a button, a dial, a slider, a touchpad, a wheel, a joystick, a keyboard, a mouse, a trackball, a camera, a microphone, a keypad, or any other such device or component. Various devices may also be connected by wireless or other such links in some embodiments. In some embodiments, a device could be controlled by a combination of visual and acoustic commands or gestures, allowing a user to control the device without having to touch the device or a physical input mechanism.

[0038] Much of the functionality used in various implementations will operate in a computing environment that may be operated by or on behalf of a service provider or entity, such as a ride-hailing company or other similar businesses. Dedicated computing resources or resources allocated as part of a multi-tenant or cloud environment may be present. The resources can utilize any number of operating systems and applications and may include any number of workstations or servers. Various implementations utilize at least a conventional network to support communication using any of a variety of commonly available protocols, such as TCP / IP or FTP, among others.As mentioned, exemplary networks include, for example, a local area network (LAN), a wide area network (WAN), a virtual private network (VPN), the internet, an intranet, and various combinations thereof. The servers used to host a service, such as a ride-sharing service, can be configured to execute programs or scripts in response to requests from user devices, such as by running one or more applications, which may be implemented as one or more scripts or programs written in any suitable programming language. The server(s) may also include one or more database servers to handle data requests and perform other such operations. Furthermore, the environment may include any number of data stores and other memory and data storage media, as discussed previously.If a system includes computerized devices, each such device may include hardware elements that can be electrically coupled via a bus or other such mechanism. Exemplary elements include, as previously discussed, at least one central processing unit (CPU) and one or more storage devices, such as disk drives, optical storage devices, and solid-state storage devices, such as random-access memory (RAM) or read-only memory (ROM), as well as removable media devices, memory cards, flash cards, etc. Such devices may also include or use one or more computer-readable storage media for storing instructions that can be executed by at least one processor of the devices.An exemplary device may also include a number of software applications, modules, services, or other elements arranged in memory, including an operating system and various application programs. It is understood that alternative embodiments may exhibit numerous variations from the one described above.

[0039] Various types of non-transient, computer-readable storage media can be used for different purposes, as discussed and suggested in this paper. This includes, for example, storing instructions or code that can be executed by at least one processor to cause the system to perform various operations. The media can correspond to any of several different media types, including volatile and non-volatile working memory, which in some implementations may be removable. Various computer-readable instructions, data structures, program modules, and other data or content can be stored on the media. Media types include, for example, RAM, DRAM, ROM, EEPROM, flash memory, solid-state memory, and other storage technologies. Other types of storage media can also be used, such as optical storage (e.g., optical memory), among other such options.These may include Blu-ray or Digital Versatile Disc (DVD) or magnetic storage media (e.g., hard disk drives or magnetic tapes). Based on the disclosures and teachings provided in this document, other possibilities and / or methods for implementing the various embodiments will be apparent to the person skilled in the art.

[0040] The description and drawings are to be considered illustrative and not restrictive. It is understood, however, that various modifications and changes may be made to them without deviating from the broader spirit and scope of the various embodiments as set forth in the patent claims. Reference symbol list 100 vehicles 102 drivers 106 seats 108 seats 110 Client computing device 112 Place of origin 114 Destination 116 sets of proposed locations 118 map 200 cards 202 starting points 204 target points 252 Starting point 254 routes 300 System 302 calculating devices 304 Network 306 Interface layer 308 Service Provider Environment 310 Promotional Machine 312 Seller Database 314 route managers 316 route memories 318 Rating generation module 320 Route optimization module 322 Price Setting Managers 324 Price Archive 326 account managers 328 User Database 330 Fleet Managers 334 assigned vehicle 400 process 410 alternate origin location 500 process 512 Place of origin 600 process 606 assigned vehicle Route 610 612 changed, predicted arrival time 614 promotional campaigns at the place of origin 700 calculating device 702 Outdoor housing 704 Display screen 706 Network or communication components 800 calculating device 802 Main Processor 804 memory 806 Data storage 808 Display element 810 performance components 812 Communication or networking component 814 Input device

Claims

A computer-implemented method comprising: Receiving, from a ride-hailing device (110; 302; 700; 800), a ride request that includes an origin and a destination; Determining a set of potential routes (254; 610) to serve the ride request; Analyzing the set of potential routes (254; 610) using a route goal function that includes a weighted combination of a set of quality metrics to generate appropriate quality ratings for the potential routes (254; 610), wherein the route goal function includes at least one rider comfort parameter and at least one operational efficiency parameter; Processing at least a subset of the potential routes (254; 610) using an optimization process to improve at least a subset of the appropriate quality ratings; Determining, by means of a route determination system, a route (254;610) in response to a ride request, wherein the route (254; 610) and a predicted time to the origin (252) are based at least partially on route history data for a variety of previously requested routes (254; 610) and the corresponding quality ratings, each previously requested route (254; 610) being associated with a corresponding origin, destination, and time period; identifying one of the vehicles (334; 606) assigned to the route (254; 610) in response to the ride request; determining, through the route determination system, a predicted time of arrival for the assigned vehicle (334; 606) to reach the origin; obtaining information for points of interest (POIs) within a distal area of ​​the origin; determining promotions (614) for the POIs at least partially based on the predicted time of arrival; and transferred to the passenger calculator (110;302; 700; 800), of computer-readable instructions relating to the promotions (614) for the POIs for display on the ride-sharing device (110; 302; 700; 800); and updating the weights of the quality metrics for the objective function based on the output of a machine learning model trained using route performance history data and current route performance data. Computer-implemented method according to claim 1, further comprising: determining a type of passenger for the ride request at least on the basis of passenger history data; and generating the promotions (614) for the POIs at least partially on the basis of the type of passenger or promotions (614) previously provided to the passenger. Computer-implemented method according to claim 2, wherein the promotions (614) previously provided to the driver (102) include previously redeemed promotions (614) and previously rejected promotions (614). Computer-implemented method according to claim 1, further comprising: predicting a delay on the route (254; 610) with respect to the predicted arrival time at the origin; calculating an estimated time to complete each promotion (614); and selecting the promotions (614) to be transferred to the passenger computing device (110; 302; 700; 800) based on whether the estimated time to complete the promotion (614) is within a period of delay. A computer-implemented method according to claim 1, further comprising: determining the route (254; 610) with an alternative origin having an alternative time to origin that is shorter than the predicted time to origin; determining promotions (614) for alternative POIs within the distal area of ​​the alternative origin; sending computer-readable instructions to the passenger computing device (110; 302; 700; 800) to display a selectable menu for the promotions (614) in exchange for the driver (102) accepting the alternative origin; receiving a selected promotion (614) in response to the selectable menu; and sending computer-readable instructions to the assigned vehicle (334; 606) to direct the assigned vehicle (334; 606) to the alternative origin. Computer-implemented method according to claim 1, further comprising: calculating a quality rating for route (254; 610); determining whether the quality rating fails to meet a quality rating threshold; and determining an alternative origin with an alternative quality rating that meets the quality rating threshold. A computer-implemented method according to claim 1, wherein determining the promotions (614) further comprises: calculating a passenger comfort rating for the origin or an alternative origin; determining whether the passenger comfort rating does not meet a passenger comfort rating threshold; and determining the promotions (614) at least partially based on the passenger comfort rating of the origin or alternative origin. Computer-implemented method according to claim 1, further comprising: calculating an estimated time to complete each promotion (614); and selecting the promotions (614) to be transferred to the passenger computing device (110; 302; 700; 800) based on whether the estimated time to complete the promotion (614) is within the predicted arrival time. A computer-implemented procedure comprising: Receiving trip requests from one or more computing devices (110; 302; 700; 800) corresponding to a future period for a geographical area, each trip request including at least an origin and a destination; Determining available vehicles (334) to serve the trip requests in the geographical area during the future period; Determining a set of route solutions at least partially based on the available vehicles (334) and the trip requests; Generating an appropriate quality score for each route solution in the set of route solutions using an objective function that includes a weighted combination of quality metrics; Identifying an alternative route solution with an alternative origin at least partially based on the alternative route (254;610), which has a corresponding quality rating higher than the corresponding quality ratings of each route solution in the set of route solutions; Analyzing map information to determine points of interest (POIs) within a proximity of the alternative point of origin; Determining promotions (614) for the POIs at least partially based on the predicted time of arrival; Transmitting computer-readable instructions relating to the promotions (614) for the POIs to the ride-hailing device (110; 302; 700; 800) for display on the ride-hailing device (110; 302; 700; 800); Receiving a response from the ride-hailing device (110; 302; 700; 800) accepting at least one of the promotions (614); and identifying an assigned vehicle (334; 606) from the available vehicles (334) to complete the alternative route (254; 610) and travel to the alternative point of origin;and updating the weights of the quality metrics for the objective function based on the output of a machine learning model trained using route performance history data and current route performance data. A computer-implemented method according to claim 9, further comprising: receiving a response from the passenger computing device (110; 302; 700; 800) rejecting the promotions (614); and performing at least one of the following: identifying a selected route solution from the set of route solutions, at least partially based on the corresponding quality rating and an assigned vehicle (334; 606) from the available vehicles (334) to execute the selected route solution; determining a second set of route solutions, at least partially based on the available vehicles (334) and the trip requirements; identifying a second alternative route (254; 610) having a second origin, with a second alternative quality rating that meets a minimum quality rating threshold;and analyzing map information to determine alternative points of interest (POIs) and alternative promotions (614) to be provided to the ride-sharing calculator (110; 302; 700; 800) for display.; Computer-implemented method according to claim 10, wherein the ride requests include at least one of: account information associated with an account of the ride-sharing device (110; 302; 700; 800), passenger capacity, cargo capacity, vehicle type, estimated time of arrival or estimated time to complete the ride requests. Computer-implemented method according to claim 11, wherein the promotions (614) or alternative promotions (614) are determined at least partially on the basis of history information associated with the account of the ride-sharing device (110; 302; 700; 800), the account information, the proximity of the POIs to the point of origin or the alternative point of origin, and an estimated time of arrival for the assigned vehicle (334; 606) to reach the point of origin or the alternative point of origin. A computer-implemented method according to claim 9, wherein the route target function includes at least one passenger comfort parameter and at least one operational efficiency parameter, wherein determining the set of routes (254; 610) further comprises: processing at least a subset of the potential route solutions using an optimization process to improve at least a subset of the corresponding quality ratings, wherein the optimization process includes one or more alternative origin locations within a proximity of the origin location; and determining the alternative route solution from the set of potential route solutions at least partially based on the corresponding quality ratings. System comprising: at least one processor; and memory (804) containing instructions which, when executed by the at least one processor, cause the system to: receive, from a ride-sharing device (110; 302; 700; 800), a ride request comprising an origin and a destination; determine a set of potential routes (254; 610) to serve the ride request; analyze the set of potential routes (254; 610) using a route goal function comprising a weighted combination of a set of quality metrics to generate appropriate quality ratings for the potential routes (254; 610), wherein the route goal function includes at least one rider comfort parameter and at least one operational efficiency parameter; process at least a subset of the potential routes (254;610) using an optimization process to improve at least a subset of the corresponding quality ratings; Determine, by means of a route determination system, a route (254; 610) in response to a ride request, wherein the route (254; 610) and a predicted time to the origin (252) are based at least partially on route history data for a plurality of previously requested routes (254; 610) and the corresponding quality ratings, each previously requested route (254; 610) being associated with a corresponding origin, destination, and time period; Identify one of the vehicles (334; 606) to be assigned to the route (254; 610) in response to the ride request; Determine, by means of the route determination system, a predicted time for the assigned vehicle (334; 606) to reach the origin;Obtaining information for points of interest (POIs) within a distal area of ​​the origin; determining promotions (614) for the POIs at least partially based on the predicted time to origin; and sending, to the passenger computing device (110; 302; 700; 800), computer-readable instructions relating to the promotions (614) for the POIs; and updating the weights of the quality metrics for the objective function based on the output of a machine learning model trained using route performance history data and current route performance data. System according to claim 14, wherein the instructions, upon execution, further cause the system to: determine a type of passenger for the ride request at least on the basis of passenger history data; and generate the promotions (614) for the POIs at least partially on the basis of the type of passenger or of at least one of previously accepted or previously rejected promotions (614) for the passenger. System according to claim 14, wherein the instructions, upon execution, further cause the system to: predict a delay on the route (254; 610) with respect to the predicted time to origin; calculate an estimated time to complete each promotion (614); and select the promotions (614) based on the estimated time to complete the promotion (614) within a period of delay. System according to claim 14, wherein the instructions, upon execution, further cause the system to: determine the route (254; 610) with an alternative origin having an alternative time to origin that is shorter than the predicted time to origin; determine promotions (614) for alternative POIs within the distal area of ​​the alternative origin; send computer-readable instructions to the passenger computing device (110; 302; 700; 800) to display a selectable menu for the promotions (614) in exchange for the driver (102) accepting the alternative origin; receive a selected promotion (614) in response to the selectable menu; and send computer-readable instructions to the assigned vehicle (334; 606) to direct the assigned vehicle (334; 606) to the alternative origin. System according to claim 14, wherein the instructions, upon execution, further cause the system to: calculate a passenger comfort rating for the origin or an alternative origin; determine whether the passenger comfort rating does not meet a passenger comfort rating threshold; and determine the promotions (614) at least partially based on the passenger comfort rating of the origin or alternative origin.

Images