Abstract: Methods, systems, apparatus and computer program code are provided for predicting a destination location of a vehicle which include processing a first and a second graph neural network to generate a prediction of a destination location.

Abstract: Aspects concern a method for controlling a transport system comprising determining historical data of a multiplicity of transport trips performed by the transport system, training a neural network to perform an invertible mapping of transport trips to latent representations to fulfil, by the distribution of the latent representations of the multiplicity of transport trips, a predetermined fitting criterion with respect to a predetermined latent representation base distribution, sampling a multiplicity of latent representations from the base distribution, mapping each of the sampled latent representations to a respective transport trip by using the trained neural network to perform the inverse of the invertible mapping to generate a multiplicity of synthetic transport trips, determining a control scheme for the transport system using the multiplicity of synthetic transport trips and controlling the transport system using the determined control scheme.

Abstract: Aspects concern a method for controlling a transport system comprising determining historical data of a multiplicity of transport trips performed by the transport system, training a neural network to perform an invertible mapping of transport trips to latent representations to fulfil, by the distribution of the latent representations of the multiplicity of transport trips, a predetermined fitting criterion with respect to a predetermined latent representation base distribution, sampling a multiplicity of latent representations from the base distribution, mapping each of the sampled latent representations to a respective transport trip by using the trained neural network to perform the inverse of the invertible mapping to generate a multiplicity of synthetic transport trips, determining a control scheme for the transport system using the multiplicity of synthetic transport trips and controlling the transport system using the determined control scheme.

Abstract: A processor device has a CPU cooperating with an input device and an output device, under control of stored instructions, and is arranged to receive service requests at the input device, assign service requests received in successive time periods to respective batches of requests; access stored service provider data to identify available service providers from among a pool of service providers; after completing the assignment of service requests to a batch, perform a matching process to endeavour to match each service request of the batch of requests to a service provider; and for each service provider to whom a match is made, output a notification of the respective potential match from the output device.

Abstract: Aspects concern a method for predicting the destination location of a vehicle, comprising processing a local preference graph of a user of the vehicle having nodes corresponding to locations visited before by the user by a first graph neural network, processing one or more of a spatial graph representing information about geographical proximity of locations, a temporal graph representing information about locations which have been visited one after another by users and the time between the visits of the locations and a preference graph representing information about locations which have been visited one after another and the frequency of visits of the locations by a second graph neural network, combining the result of the processing by the first graph neural network and the result of the processing by the second graph neural network by at least one neural network layer and using the output of the at least one neural network layer as prediction of the destination location.

Abstract: A system for predicting a destination location may include one or more processors and a memory having instructions stored therein. The one or more processors may use at least one recurrent neural network to: process spatial data which may include a first set of information about origin locations and destination locations; process temporal data which may include a second set of information about times at the origin locations and the destination locations; determine hidden state data based on the spatial data and the temporal data, wherein the hidden state data may include data on origin-destination relationships; receive a current input data from a user, wherein the current input data may include an identity of the user and the current origin location of the user; and predict the destination location based on the hidden state data and the current input data.

Abstract: Aspects concern a method for controlling a transport system comprising determining historical data of a multiplicity of transport trips performed by the transport system, training a neural network to perform an invertible mapping of transport trips to latent representations to fulfil, by the distribution of the latent representations of the multiplicity of transport trips, a predetermined fitting criterion with respect to a predetermined latent representation base distribution, sampling a multiplicity of latent representations from the base distribution, mapping each of the sampled latent representations to a respective transport trip by using the trained neural network to perform the inverse of the invertible mapping to generate a multiplicity of synthetic transport trips, determining a control scheme for the transport system using the multiplicity of synthetic transport trips and controlling the transport system using the determined control scheme.

Abstract: Disclosed are systems and methods for predicting the destination location of a vehicle by processing a local preference graph of a user of the vehicle having nodes corresponding to locations visited by the user. In some embodiments, information about locations which have been visited one after another by users and the time between visits of the locations are also used in the prediction.

Abstract: A server configured to determine a fee surge for an on-demand service may include processor(s) that may perform the following upon receiving each request for the service: determine, from the request, a location where the service is required; determine, using a map from the memory, a supply area around the location; determine the surge based on the supply area; compare the surge with a surge lower bound for the supply area and set it as this bound if it is lower than this bound; communicate the surge to the respective requestor computing device and a provider computing device; and allocate the associated service provider to fulfil the request if both the service requestor and service provider accept the surge. The surge lower bound may be based on historical allocation rates, each representing a number of fulfilled requests at a fee surge during historical time interval(s) in a characteristically similar area.

Abstract: A communications server apparatus and method for forecasting demand for services is provided, the server managing the provision of services based on received queries. Data records comprising data instances of service queries within a past time period are received. Modified data records are derived by modifying a value of a feature of one or more of the data instances to a standardized value. The feature is a parameter relating to an offer, by the server, to provide a service in response to a query. The data instances of the modified data records are input into a classification model, wherein the classification model predicts whether or not a service booking will be made. Classification predictions are output by the classification model and an estimate of potential demand for the past time period is determined based on the classification predictions, for improved forecasting of demand for services.

Abstract: Aspects concern a method for predicting the destination location of a vehicle, comprising processing a local preference graph of a user of the vehicle having nodes corresponding to locations visited before by the user by a first graph neural network, processing one or more of a spatial graph representing information about geographical proximity of locations, a temporal graph representing information about locations which have been visited one after another by users and the time between the visits of the locations and a preference graph representing information about locations which have been visited one after another and the frequency of visits of the locations by a second graph neural network, combining the result of the processing by the first graph neural network and the result of the processing by the second graph neural network by at least one neural network layer and using the output of the at least one neural network layer as prediction of the destination location.

Abstract: A server configured to determine an advanced booking fee for an advance booking is disclosed. The server may include one or more processor(s) and a memory having instructions stored therein. The instructions when executed by the one or more processor(s), may cause the one or more processor(s) to receive an advanced booking enquiry over a communication network from a user device for a predetermined transportation service at a predetermined future time from a service provider. The one or more processor(s) may determine a call option price based on a difference between a predetermined fare price of the advance booking and a potential future trip price. The one or more processor(s) may determine an earning adjustment for the service provider to provide the predetermined transportation service at the predetermined future time.

Abstract: Aspects concern a demand notification device (110), comprising: a determining unit (122) configured to determine a quantity of a demand of a transport service for a plurality of users (102, 104, 106) having a predetermined area as destination in a first time period, the quantity of the demand indicating how many users of the plurality of users are determined to desire to travel into the predetermined area; and wherein the determining unit (122) is further configured to determine a real space service demand from a plurality of users (102, 104, 106) to be fulfilled in the predetermined area in a second time period, wherein the real space service is provided by a service provider (120); an analysis unit (124) configured to determine a predicted real space service demand in a third time period for the predetermined area based on the quantity of users (102, 104, 106) in the predetermined area at the first time period and the demand of real space service in the predetermined area in the second time period and furth

Abstract: A processor device has a CPU cooperating with an input device and an output device, under control of stored instructions, and is arranged to receive service requests at the input device, assign service requests received in successive time periods to respective batches of requests; access stored service provider data to identify available service providers from among a pool of service providers; after completing the assignment of service requests to a batch, perform a matching process to endeavour to match each service request of the batch of requests to a service provider; and for each service provider to whom a match is made, output a notification of the respective potential match from the output device.

Abstract: Server apparatus comprising a processor and a memory, the server apparatus being configured, under control of the processor, to execute instructions stored in the memory: to establish a route data record comprising data indicative of plural routes, each said route being between a start location and a respective destination, whereby each respective destination has at least one route to it from said start location; to process the route data record to determine a zone data record indicative of zones making up each route; and to combine the zone data record with a prediction data record to score the route data making up each said route to each destination based upon a prediction of the probability of a job or expected revenue in each zone making up that route.

Abstract: Example embodiments relate generally to methods, systems, and devices for managing service providers and service requests. The method includes, for each identified geographical area, deriving a service request forecast and service provider forecast for a particular upcoming time period. The method includes, for each identified geographical area, determining whether the geographical area will be in an over-supply state during the particular upcoming time period. The method includes, for each identified geographical area determined to be in the over-supply state during the particular upcoming time period: determining a quantity M of available service providers; selecting at least M available service providers in the geographical area; and providing a notification to only the selected available service providers. Each notification may include a message to move out of the geographical area and into a particular location in another geographical area.

Abstract: A processor device has a CPU cooperating with an input device and an output device, under control of stored instructions, and is arranged to receive service requests at the input device, assign service requests received in successive time periods to respective batches of requests; access stored service provider data to identify available service providers from among a pool of service providers; after completing the assignment of service requests to a batch, perform a matching process to endeavour to match each service request of the batch of requests to a service provider; and for each service provider to whom a match is made, output a notification of the respective potential match from the output device.