| CPC G06Q 50/40 (2024.01) [G06F 16/2379 (2019.01); G06Q 10/06315 (2013.01); G08G 1/202 (2013.01)] | 7 Claims |
|
1. A processor implemented method, comprising:
receiving (202), from a plurality of users, a plurality of incoming travel requests, each incoming travel request from the plurality of incoming travel requests corresponds to a user from the plurality of users;
querying a system database (204), based on the plurality of incoming travel requests, to determine at least one of (i) whether one or more incoming travel requests are new travel requests, and (ii) inconsistencies in information comprised in the system database and to identify one or more vehicles and obtain a list comprising information corresponding to the one or more vehicles, wherein the system database comprises information pertaining to a facility of an organization, wherein the system database comprises information of each of users, vehicles, and enterprise management unit, wherein the information of each user comprises personal details of users, shift timings, reckoner distance of location of each user with respect to the facility, location of each user, and gender of each user, wherein the vehicle information comprises a number of available vehicles, type of the available vehicles, type of vehicles used for a shift pertaining to one facility, and capacity of each of the available vehicles, and wherein the information of the enterprise management unit comprises a list of users requesting for transportation, a previous trip history, combining patterns of locations during previously executed trips, and trip sheets generated;
implementing a self-learning based mechanism for vehicle utilization of the one or more vehicles to dynamically assign the plurality of users to the one or more vehicles, wherein a custom machine learning based model is utilized for dynamic assignment of the plurality of users to the one or more vehicles;
learning by the custom machine learning model, details available in a system database as past history for learning one or more previously executed trips, the system database comprising information and the information including at least one user details, shift timing, vehicle types, vehicle capacity, reckoner distance of location of the plurality of users and a number of users from the plurality of users along with localities that were previously clubbed together;
generating by the self-learning based mechanism locality pairings and clubbing patterns once training is initiated on data available from the system database, wherein the self-learning based mechanism learns vehicle types used for a given shift for one or more facilities and learnings are stored in a separate database, the learnings being used to analyze rosters in real-time for at least one shift for the one or more facilities;
learning one or more previously executed trips to analyze clubbing of the localities and identifying a particular time when the vehicle is not allowed to go in a certain locality of the one or more localities, wherein the self-learning based mechanism is trained with information provided by the separate database, the information being at least one of vehicle constraints, and social constraints;
utilizing an associative classification mechanism based on the learnings to find and extract associations from the list of previously executed trips, the vehicle constraints and the social constraints and eliminating one or more exceptions from the list of previously executed trips;
dynamically updating the system database by learning (i) information pertaining to one or more new requests, (ii) the inconsistencies in information comprised in the system database, and (iii) information associated with one or more underutilized vehicles;
identifying (206) a plurality of locations being accessible by the identified one or more vehicles with highest capacity;
identifying (208), (i) a first location from the plurality of locations and (ii) a first set of users from the plurality of users based on the first location, wherein the first location corresponds to farthest location and is determined based on a distance from the facility of the organization;
identifying (210), (i) one or more locations from the plurality of locations and (ii) a second set of users from the plurality of users, such that the one or more locations being identified are based on frequency of the one or more locations that are previously combined with the first location to obtain an optimal set of locations, wherein the frequency of the one or more locations is used to find associations between previous clubbed locations along with the users going to the previous clubbed locations;
dynamically allocating (212) each user from the first set and second set of users to the identified one or more vehicles based on (i) one or more social constraints, and (ii) one or more constraints associated with the identified one or more vehicles, wherein the one or more social constraints comprises:
(i) at least one user being identified as a female user having at least one of (a) last drop, and (b) a first pick up;
(ii) at least one user being identified as a male user from the first set of users or the second set of users for accompanying with the female user having the last drop; and
(iii) at least one security personnel being identified in addition to number of users for an identified vehicle, wherein the at least one security personnel is identified when the male user is not available, and
wherein the one or more constraints associated with the identified one or more vehicles comprise:
restrictive access to one or more locations based on size of the identified one or more vehicles; and
pick up and drop timing-based selection of one or more vehicles, from and to the one or more locations; and
generating (214), a trip schedule for the identified one or more vehicles, based on the allocation of the first set of users and the second set of users into a corresponding vehicle from the identified one or more vehicles, thereby providing self-learning capability which ensures dynamic assignment of the first set of users and the second set of users to the identified one or more vehicles and vehicle utilization in real-time, wherein the trip schedule is generated by adaptively learning an association of the one or more locations with the identified first location and dynamically assigning the first set of users and the second set of users to the identified one or more vehicles based on the association.
|