Artificial intelligence (AI) based systems and methods for analyzing order data to generate a driver logistics prediction value

1 . An artificial intelligence (AI) based system configured to analyze order data to generate a driver logistics prediction value, the system comprising:

a logistics server configured to ingest a set of order data from one or more order data sources, the logistics server comprising computing instructions configured to execute on one or more processors of the logistics server, wherein the logistics server is communicatively coupled to a dispatch application (app); and

a dispatch learning model, accessible by the logistics server, that is trained using (i) logistics data regarding thousands of instances of training order data of respective orders including in-gate times and outgate times associated with one or more locations, (ii) respective location data of respective drivers, (iii) respective delivery routes of respective drivers, and (iv) a plurality of formats of the training order data, (v) driver details for a plurality of drivers including at least one of: (a) a driver location or (b) a driver credential value, and (vi) driver vehicle details associated with the plurality of drivers including at least one of: (a) a fuel level or (b) a fuel consumption rate to accurately predict the driver logistics prediction value, the dispatch learning model being trained to output (a) one or more driver logistics prediction values corresponding to the training order data of the respective orders and (b) the training order data in a universal format

wherein the dispatch learning model has at least one of: (i) a depth value associated with layers of the dispatch learning model or (ii) a width value associated with nodes in the layers of the dispatch learning model, the depth value and the width value being configured to decrease machine resources of the logistics server consumed during execution of the dispatch learning model, wherein the dispatch learning model decreases computational resources by reducing at least one of the depth value or the width value dimensionality requirement of machine resources of the logistics server, thereby freeing up computational resources of the logistics server,

wherein the computing instructions, when executed by the one or more processors of the logistics server, cause the one or more processors of the logistics server to:

receive, by the logistics server, a dispatch intake form in a first format from the one or more order data sources corresponding to an order, wherein the first format is related to a transmission method used by a respective order data source of the one or more order data sources that submitted the dispatch intake form to the logistics server,

extract, using a character recognition routine, order data from the dispatch intake form, wherein the order data includes a pick-up yard, a destination location, an order status, and a priority value,

standardize, by the dispatch learning model, the order data from the first format into the universal format,

receive, from the dispatch app, (i) a respective location, (ii) a respective delivery route of each respective driver with a respective device executing the dispatch app, and (iii) a delivery status of the driver,

analyze, by the dispatch learning model, (i) the order data in the universal format, (ii) historical in-gate times associated with the pick-up yard that are stored in a historical record that includes in-gate times and outgate times associated with one or more locations, (iii) historical outgate times associated with the pick-up yard that are stored in the historical record, (iv) the respective location of each respective driver, and (v) the respective delivery routes of each respective driver to:

generate, via a mapping service and for each respective driver based on the respective locations, the delivery status, and the respective delivery routes of the respective drivers, an optimal route configured to minimize time required for each driver to complete their respective delivery routes and travel to the pick-up yard, and

generate a driver logistics prediction value corresponding to the order data based on a first optimal route for a driver of the respective drivers, the first optimal route having a minimal total time for the driver to arrive at the pick-up yard relative to other optimal routes generated via the mapping service

wherein the dispatch learning model generates the optimal route by calculating estimated arrival times based on the historical in-gate times and the historical outgate times stored in the historical record, thereby reducing computational iterations required to determine driver assignments,

generate, based on the driver logistics prediction value, a driver determination that identifies the driver to fulfill the order,

push, by the logistics server, the driver determination to the dispatch app of the driver, and wherein, responsive to receiving the driver determination, the computing instructions cause the dispatch app to display a first graphical user interface (GUI) including the driver determination,

receive a first updated order status from the dispatch app of the driver indicating that the order is assigned to the driver, and

update the dispatch learning model using the driver logistics prediction value to improve the accuracy of a subsequent driver logistics prediction value output by the dispatch learning model.

2 . The AI based system of claim 1 , wherein the computing instructions further cause the one or more processors of the logistics server to:

determine an order status corresponding to the order based on the order data, and

responsive to determining that the order status indicates a grounded load, analyze, by the dispatch learning model, the order data to generate the driver logistics prediction value corresponding to the order data.

3 . The AI based system of claim 1 , wherein the order data is included in a plurality of dispatch intake forms in a plurality of formats, and the computing instructions further cause the one or more processors of the logistics server to:

extract respective order data from each of the plurality of dispatch intake forms,

standardize the respective order data into the universal format, and

analyze, by the dispatch learning model, each of the respective order data to generate a respective driver logistics prediction value for each of the respective order data.

4 . The AI based system of claim 1 , wherein the driver determination includes at least a portion of the order data.

5 . The AI based system of claim 1 , wherein the one or more order data sources includes at least one of (1) a website, (2) an email message, or (3) a spreadsheet.

6 . The AI based system of claim 5 , wherein the one or more order data sources includes the website, and the website includes a website extension configured to communicate with the logistics server to transfer the order data from the website to the logistics server.

7 . The AI based system of claim 1 , wherein the order data is a plurality of order data corresponding to a plurality of orders, each respective order data includes a priority value, and the computing instructions further cause the one or more processors of the logistics server to:

generate a ranked list of the plurality of orders based on the priority value included in each respective order data,

receive, from the one or more order data sources, a new order data corresponding to a new order, the new order data including a high priority value, and

re-rank the ranked list of the plurality of orders based on the high priority value of the new order data.

8 . The AI based system of claim 1 , wherein the computing instructions further cause the one or more processors of the logistics server to:

receive (i) the dispatch intake form in the first format from the one or more order data sources, (ii) driver details for a plurality of drivers including at least one of: (a) a driver location or (b) a driver credential value, and (iii) driver vehicle details associated with the plurality of drivers including at least one of: (a) a fuel level or (b) a fuel consumption rate; and

analyze, by the dispatch learning model, (i) the order data in the universal format, (ii) the driver details, and (iii) the driver vehicle details to generate the driver logistics prediction value corresponding to the order data, wherein the dispatch learning model is trained or provided with the logistics data regarding order data of respective orders, wherein the dispatch learning model is configured to output the one or more driver logistics prediction values corresponding to the order data of the respective orders, and wherein the driver logistics prediction value represents a time for a driver to complete the order based on (i) the driver location and (ii) time loss resulting from the driver refueling a vehicle corresponding to the driver vehicle details.

9 . The AI based system of claim 1 , wherein both the pick-up yard and the destination location are railway terminals, and the computing instructions further cause the one or more processors of the logistics server to:

determine an order status corresponding to the order based on the order data, wherein the order status includes at least one of: (i) an arrived status indicating a train transporting the order has arrived at the pick-up yard, (ii) a placed status indicating the order is ready for offloading from the train, or (iii) a grounded status indicating the order is ready for transportation from the pick-up yard to the destination location;

retrieve, by accessing a plurality of dispatch applications (apps) of a plurality of drivers via an application programming interface (API), location information corresponding to a plurality of drivers;

responsive to determining that the order status indicates a grounded load, analyze, by the dispatch learning model, (i) the order data in the universal format and (ii) the location information to generate the driver logistics prediction value corresponding to the order data;

push the driver determination to a respective dispatch app of the driver, wherein, responsive to receiving the driver determination, the respective dispatch app displays the first GUI depicting the driver determination; and

responsive to the driver accepting the order via the driver determination, receive the updated order status from the respective dispatch app of the driver indicating that the order is assigned to the driver.

10 . The AI based system of claim 1 , wherein the computing instructions further cause the one or more processors of the logistics server to:

receive (i) the dispatch intake form in the first format from the one or more order data sources, (ii) driver details for a plurality of drivers including at least one of: (a) a driver location or (b) a driver credential value, and (iii) driver vehicle details associated with the plurality of drivers including at least one of: (a) a fuel level or (b) a fuel consumption rate;

extract, by executing the character recognition routine, the order data from the dispatch intake form, wherein both the pick-up yard and the destination location are railway terminals;

determine an order status corresponding to the order based on the order data, wherein the order status includes at least one of: (i) an arrived status indicating a train transporting the order has arrived at the pick-up yard, (ii) a placed status indicating the order is ready for offloading from the train, or (iii) a grounded status indicating the order is ready for transportation from the pick-up yard to the destination location;

retrieve, by accessing a plurality of dispatch applications (apps) of a plurality of drivers via an application programming interface (API), location information corresponding to a plurality of drivers;

analyze, by the dispatch learning model, (i) the order data in the universal format, (ii) the driver details, (iii) the driver vehicle details, and (iv) the location information to generate the driver logistics prediction value corresponding to the order data, wherein the dispatch learning model is trained or provided with the logistics data regarding order data of respective orders, wherein the dispatch learning model is configured to output the one or more driver logistics prediction values corresponding to the order data of the respective orders, and wherein the driver logistics prediction value represents a time for a driver to complete the order based on (i) the driver location and (ii) time loss resulting from the driver refueling a vehicle corresponding to the driver vehicle details;

push the driver determination to a respective dispatch app of the driver, wherein, responsive to receiving the driver determination, the respective dispatch app displays the first GUI depicting the driver determination; and

responsive to the driver accepting the order via the driver determination, receive the updated order status from the respective dispatch app of the driver indicating that the order is assigned to the driver.

11 . The AI based system of claim 1 , wherein the driver proceeds to deliver the order using the optimal route.

12 . The AI based system of claim 11 , wherein the computing instructions further cause the one or more processors of the logistics server to:

receive a second updated order status from the dispatch app of the driver indicating a completion of the order by the driver.

13 . The AI based system of claim 1 , wherein the driver delivers the order before a cut-off time to prevent a corresponding delay.

14 . The AI based system of claim 1 , wherein the computing instructions further cause the one or more processors of the logistics server to:

cause the dispatch app of the driver to display a second GUI including a first portion of the driver determination, the first portion comprising a scheduled order section that displays accepted orders of the driver on a particular date including the order,

track (i) an in-gate time of the driver arriving at the pick-up yard on the particular date and (ii) an outgate time of the driver departing the pick-up yard,

include (i) the in-gate time of the driver arriving at the pick-up yard on the particular date and (ii) the outgate time as entries in the historical record to maintain the historical record,

cause the dispatch app of the driver to display a third GUI including a second portion of the driver determination, the second portion comprising an order travel details section and a selectable navigation button, the order travel details section indicating the pick-up yard of the order and the destination location of the order, and

responsive to receiving a selection indication of the selectable navigation button from the driver, causing the dispatch app of the driver to render a graphical display of a second optimal route associated with the driver to provide the driver with travel instructions from the pick-up yard to the destination location.

15 . An artificial intelligence (AI) based method for analyzing order data to generate a driver logistics prediction value, the method comprising:

receiving, at one or more processors of a logistics server, a dispatch intake form in a first format from one or more order data sources corresponding to an order, wherein the first format is related to a transmission method used by a respective order data source of the one or more order data sources that submitted the dispatch intake form to the logistics server;

extracting, by the one or more processors executing a character recognition routine, order data from the dispatch intake form, wherein the order data includes a pick-up yard, a destination location, an order status, and a priority value;

standardizing, by the one or more processors executing a dispatch learning model, the order data from the first format into a universal format;

receiving, from a dispatch application (app) communicatively coupled to the logistics server, (i) a respective location, (ii) a respective delivery route of each respective driver with a respective device executing the dispatch app, (iii) a delivery status of the driver;

analyzing, by the one or more processors executing the dispatch learning model, (i) the order data in the universal format, (ii) historical in-gate times associated with the pick-up yard that are stored in a historical record that includes in-gate times and outgate times associated with one or more locations, (iii) historical outgate times associated with the pick-up yard that are stored in the historical record, (iv) the respective location of each respective driver, and (v) the respective delivery routes of each respective driver to:

generate, via a mapping service and for each respective driver based on the respective locations, the delivery status, and the respective delivery routes of the respective drivers, an optimal route configured to minimize time required for each driver to complete their respective delivery routes and travel to the pick-up yard, and

generate a driver logistics prediction value corresponding to the order data based on a first optimal route for a driver of the respective drivers, the first optimal route having a minimal total time for the driver to arrive at the pick-up yard relative to other optimal routes generated via the mapping service

wherein the dispatch learning model generates the optimal route by calculating estimated arrival times based on the historical in-gate times and the historical outgate times stored in the historical record, thereby reducing computational iterations required to determine driver assignments,

wherein the dispatch learning model is trained using (i) logistics data regarding thousands of instances of training order data of respective orders including in-gate times and outgate times associated with one or more locations, (ii) respective location data of respective drivers, (iii) respective delivery routes of respective drivers, (iv) a plurality of formats of the training order data, (v) driver details for a plurality of drivers including at least one of: (a) a driver location or (b) a driver credential value, and (vi) driver vehicle details associated with the plurality of drivers including at least one of: (a) a fuel level or (b) a fuel consumption rate to accurately predict the driver logistics prediction value, the dispatch learning model being trained to output (a) one or more driver logistics prediction values corresponding to the training order data of the respective orders and (b) the training order data in the universal format

wherein the dispatch learning model has at least one of: (i) a depth value associated with layers of the dispatch learning model or (ii) a width value associated with nodes in the layers of the dispatch learning model, the depth value and the width value being configured to decrease machine resources of the logistics server consumed during execution of the dispatch learning model, wherein the dispatch learning model decreases computational resources by reducing at least one of the depth value or the width value dimensionality requirement of machine resources of the logistics server, thereby freeing up computational resources of the logistics server;

generating, by the one or more processors, a driver determination that identifies the driver to fulfill the order based on the driver logistics prediction value;

pushing, by the one or more processors, the driver determination to the dispatch of the driver, wherein, responsive to receiving the driver determination, the dispatch app displays a first graphical user interface (GUI) depicting the driver determination;

receiving, at the one or more processors, an updated order status from the dispatch app of the driver indicating that the order is assigned to the driver; and

updating, by the one or more processors, the dispatch learning model using the driver logistics prediction value to improve the accuracy of a subsequent driver logistics prediction value output by the dispatch learning model.

16 . The AI based method of claim 15 , further comprising:

determining, by the one or more processors, an order status corresponding to the order based on the order data; and

responsive to determining that the order status indicates a grounded load, analyzing, by the dispatch learning model, the order data to generate the driver logistics prediction value corresponding to the order data.

17 . The AI based method of claim 15 , wherein the order data is a plurality of dispatch intake forms in a plurality of formats, and the method further comprises:

extracting, by the one or more processors, respective order data from each of the plurality of dispatch intake forms;

standardizing, by the one or more processors, the respective order data into the universal format; and

analyzing, by the dispatch learning model, each of the respective order data to generate a respective driver logistics prediction value for each of the respective order data.

18 . The AI based method of claim 15 , wherein the driver determination includes at least a portion of the order data.

19 . The AI based method of claim 15 , wherein the one or more order data sources includes at least one of (1) a website, (2) an email message, or (3) a spreadsheet.

20 . The AI based method of claim 19 , wherein the one or more order data sources includes the website, and the website includes a website extension configured to communicate with the logistics server to transfer the order data from the website to the logistics server.

21 . The AI based method of claim 15 , wherein the order data is a plurality of order data corresponding to a plurality of orders, each respective order data includes a priority value, and the method further comprises:

generating, by the one or more processors, a ranked list of the plurality of orders based on the priority value included in each respective order data;

receiving, from the one or more order data sources, a new order data corresponding to a new order, the new order data including a high priority value; and

re-ranking, by the one or more processors, the ranked list of the plurality of orders based on the high priority value of the new order data.

22 . The AI based method of claim 15 , further comprising:

receiving, at the one or more processors, (i) the dispatch intake form in the first format from the one or more order data sources, (ii) driver details for a plurality of drivers including at least one of: (a) a driver location or (b) a driver credential value, and (iii) driver vehicle details associated with the plurality of drivers including at least one of: (a) a fuel level or (b) a fuel consumption rate; and

analyzing, by the dispatch learning model, (i) the order data in the universal format, (ii) the driver details, and (iii) the driver vehicle details to generate the driver logistics prediction value corresponding to the order data, wherein the dispatch learning model is trained or provided with the logistics data regarding order data of respective orders, wherein the dispatch learning model is configured to output the one or more driver logistics prediction values corresponding to the order data of the respective orders, and wherein the driver logistics prediction value represents a time for a driver to complete the order based on (i) the driver location and (ii) time loss resulting from the driver refueling a vehicle corresponding to the driver vehicle details.

23 . The AI based method of claim 15 , wherein both the pick-up yard and the destination location are railway terminals, and the method further comprises:

determining, by the one or more processors, an order status corresponding to the order based on the order data, wherein the order status includes at least one of: (i) an arrived status indicating a train transporting the order has arrived at the pick-up yard, (ii) a placed status indicating the order is ready for offloading from the train, or (iii) a grounded status indicating the order is ready for transportation from the pick-up yard to the destination location;

retrieving, by the one or more processors accessing a plurality of dispatch applications (apps) of a plurality of drivers via an application programming interface (API), location information corresponding to a plurality of drivers;

responsive to determining that the order status indicates a grounded load, analyzing, by the dispatch learning model, (i) the order data in the universal format and (ii) the location information to generate the driver logistics prediction value corresponding to the order data;

pushing, by the one or more processors, the driver determination to a respective dispatch app of the driver, wherein, responsive to receiving the driver determination, the respective dispatch app displays the graphical user interface (GUI) depicting the driver determination; and

responsive to the driver accepting the order via the driver determination, receiving, at the one or more processors, the updated order status from the respective dispatch app of the driver indicating that the order is assigned to the driver.

24 . The AI based method of claim 15 , further comprising:

receiving, at the one or more processors, (i) the dispatch intake form in the first format from the one or more order data sources, (ii) driver details for a plurality of drivers including at least one of: (a) a driver location or (b) a driver credential value, and (iii) driver vehicle details associated with the plurality of drivers including at least one of: (a) a fuel level or (b) a fuel consumption rate;

extracting, by the one or more processors executing the character recognition routine, the order data from the dispatch intake form, wherein both the pick-up yard and the destination location are railway terminals;

determining, by the one or more processors, an order status corresponding to the order based on the order data, wherein the order status includes at least one of: (i) an arrived status indicating a train transporting the order has arrived at the pick-up yard, (ii) a placed status indicating the order is ready for offloading from the train, or (iii) a grounded status indicating the order is ready for transportation from the pick-up yard to the destination location;

retrieving, by the one or more processors accessing a plurality of dispatch applications (apps) of a plurality of drivers via an application programming interface (API), location information corresponding to a plurality of drivers;

analyzing, by the dispatch learning model, (i) the order data in the universal format, (ii) the driver details, (iii) the driver vehicle details, and (iv) the location information to generate the driver logistics prediction value corresponding to the order data, wherein the dispatch learning model is trained or provided with the logistics data regarding order data of respective orders, wherein the dispatch learning model is configured to output the one or more driver logistics prediction values corresponding to the order data of the respective orders, and wherein the driver logistics prediction value represents a time for a driver to complete the order based on (i) the driver location and (ii) time loss resulting from the driver refueling a vehicle corresponding to the driver vehicle details;

pushing, by the one or more processors, the driver determination to a respective dispatch app of the driver, wherein, responsive to receiving the driver determination, the respective dispatch app displays the graphical user interface (GUI) depicting the driver determination; and

responsive to the driver accepting the order via the driver determination, receiving, at the one or more processors, the updated order status from the respective dispatch app of the driver indicating that the order is assigned to the driver.

25 . The AI based method of claim 15 , further comprising:

updating, by the one or more processors, the dispatch learning model using the driver logistics prediction value to improve the accuracy of a subsequent driver logistics prediction value output by the dispatch learning model.

26 . The AI based method of claim 15 , wherein the driver proceeds to deliver the order using the optimal route.

27 . The AI based method of claim 26 , further comprising:

receiving, by the one or more processors, a second updated order status from the dispatch app of the driver indicating a completion of the order by the driver.

28 . The AI based method of claim 15 , wherein the driver delivers the order before a cut-off time to prevent a corresponding delay.

29 . The AI based method of claim 15 , further comprising:

causing, by the one or more processors, the dispatch app of the driver to display a second GUI including a first portion of the driver determination, the first portion comprising a scheduled order section that displays accepted orders of the driver on a particular date including the order,

tracking, by the one or more processors, (i) an in-gate time of the driver arriving at the pick-up yard on the particular date and (ii) an outgate time of the driver departing the pick-up yard,

including, by the one or more processors, (i) the in-gate time of the driver arriving at the pick-up yard on the particular date and (ii) the outgate time as entries in the historical record to maintain the historical record,

causing, by the one or more processors, the dispatch app of the driver to display a third GUI including a second portion of the driver determination, the second portion comprising an order travel details section and a selectable navigation button, the order travel details section indicating the pick-up yard of the order and the destination location of the order, and

responsive to receiving a selection indication of the selectable navigation button from the driver, causing, by the one or more processors, the dispatch app of the driver to render a graphical display of a second optimal route associated with the driver to provide the driver with travel instructions from the pick-up yard to the destination location.

30 . A tangible, non-transitory computer-readable medium storing instructions for analyzing order data to generate a driver logistics prediction value, that when executed by one or more processors cause the one or more processors to:

receive a dispatch intake form in a first format from one or more order data sources corresponding to an order, wherein the first format is related to a transmission method used by a respective order data source of the one or more order data sources that submitted the dispatch intake form to a logistics server;

extract, using a character recognition routine, order data from the dispatch intake form, wherein the order data includes a pick-up yard, a destination location, an order status, and a priority value;

standardize, by a dispatch learning model, the order data from the first format into a universal format;

receive, from a dispatch application (app) communicatively coupled to the logistics server, (i) a respective location, (ii) a respective delivery route of each respective driver with a respective device executing the dispatch app, (iii) a delivery status of the driver;

analyze, by the dispatch learning model, (i) the order data in the universal format, (ii) historical in-gate times associated with the pick-up yard that are stored in a historical record that includes in-gate times and outgate times associated with one or more locations, (iii) historical outgate times associated with the pick-up yard that are stored in the historical record, (iv) the respective location of each respective driver, and (v) the respective delivery routes of each respective driver to:

generate, via a mapping service and for each respective driver based on the respective locations, the delivery status, and the respective delivery routes of the respective drivers, an optimal route configured to minimize time required for each driver to complete their respective delivery routes and travel to the pick-up yard, and

generate a driver logistics prediction value corresponding to the order data based on a first optimal route for a driver of the respective drivers, the first optimal route having a minimal total time for the driver to arrive at the pick-up yard relative to other optimal routes generated via the mapping service

wherein the dispatch learning model generates the optimal route by calculating estimated arrival times based on the historical in-gate times and the historical outgate times stored in the historical record, thereby reducing computational iterations required to determine driver assignments,

wherein the dispatch learning model is trained using (i) logistics data regarding thousands of instances of training order data of respective orders including in-gate times and outgate times associated with one or more locations, (ii) respective location data of respective drivers, (iii) respective delivery routes of respective drivers, (iv) a plurality of formats of the training order data, (v) driver details for a plurality of drivers including at least one of: (a) a driver location or (b) a driver credential value, and (vi) driver vehicle details associated with the plurality of drivers including at least one of: (a) a fuel level or (b) a fuel consumption rate to accurately predict the driver logistics prediction value, the dispatch learning model being trained to output (a) one or more driver logistics prediction values corresponding to the training order data of the respective orders and (b) the training order data in the universal format

wherein the dispatch learning model has at least one of: (i) a depth value associated with layers of the dispatch learning model or (ii) a width value associated with nodes in the layers of the dispatch learning model, the depth value and the width value being configured to decrease machine resources of the logistics server consumed during execution of the dispatch learning model, wherein the dispatch learning model decreases computational resources by reducing at least one of the depth value or the width value dimensionality requirement of machine resources of the logistics server, thereby freeing up computational resources of the logistics server;

generate a driver determination that identifies the driver to fulfill the order based on the driver logistics prediction value;

push the driver determination to the dispatch of the driver, wherein, responsive to receiving the driver determination, the dispatch app displays a first graphical user interface (GUI) depicting the driver determination;

receive an updated order status from the dispatch app of the driver indicating that the order is assigned to the driver; and

update the dispatch learning model using the driver logistics prediction value to improve the accuracy of a subsequent driver logistics prediction value output by the dispatch learning model.

31 . The tangible, non-transitory computer-readable medium of claim 30 , wherein the instructions further cause the one or more processors to:

determine an order status corresponding to the order based on the order data; and

responsive to determining that the order status indicates a grounded load, analyze, by the dispatch learning model, the order data to generate the driver logistics prediction value corresponding to the order data.

32 . The tangible, non-transitory computer-readable medium of claim 30 , wherein the order data is a plurality of dispatch intake forms in a plurality of formats, and the instructions further cause the one or more processors to:

extract respective order data from each of the plurality of dispatch intake forms;

standardize the respective order data into the universal format; and

analyze, by the dispatch learning model, each of the respective order data to generate a respective driver logistics prediction value for each of the respective order data.

33 . The tangible, non-transitory computer-readable medium of claim 30 , wherein the instructions further cause the one or more processors to:

receive (i) the dispatch intake form in the first format from the one or more order data sources, (ii) driver details for a plurality of drivers including at least one of: (a) a driver location or (b) a driver credential value, and (iii) driver vehicle details associated with the plurality of drivers including at least one of: (a) a fuel level or (b) a fuel consumption rate; and

analyze, by the dispatch learning model, (i) the order data in the universal format, (ii) the driver details, and (iii) the driver vehicle details to generate the driver logistics prediction value corresponding to the order data, wherein the dispatch learning model is trained or provided with the logistics data regarding order data of respective orders, wherein the dispatch learning model is configured to output the one or more driver logistics prediction values corresponding to the order data of the respective orders, and wherein the driver logistics prediction value represents a time for a driver to complete the order based on (i) the driver location and (ii) time loss resulting from the driver refueling a vehicle corresponding to the driver vehicle details.

34 . The tangible, non-transitory computer-readable medium of claim 30 , wherein both the pick-up yard and the destination location are railway terminals, and the instructions further cause the one or more processors to:

determine an order status corresponding to the order based on the order data, wherein the order status includes at least one of: (i) an arrived status indicating a train transporting the order has arrived at the pick-up yard, (ii) a placed status indicating the order is ready for offloading from the train, or (iii) a grounded status indicating the order is ready for transportation from the pick-up yard to the destination location;

retrieve, by accessing a plurality of dispatch applications (apps) of a plurality of drivers via an application programming interface (API), location information corresponding to a plurality of drivers;

responsive to determining that the order status indicates a grounded load, analyze, by the dispatch learning model, (i) the order data in the universal format and (ii) the location information to generate the driver logistics prediction value corresponding to the order data;

push the driver determination to a respective dispatch app of the driver, wherein, responsive to receiving the driver determination, the respective dispatch app displays the graphical user interface (GUI) depicting the driver determination; and

responsive to the driver accepting the order via the driver determination, receive the updated order status from the respective dispatch app of the driver indicating that the order is assigned to the driver.

35 . The tangible, non-transitory computer-readable medium of claim 30 , wherein the instructions further cause the one or more processors to:

receive (i) the dispatch intake form in the first format from the one or more order data sources, (ii) driver details for a plurality of drivers including at least one of: (a) a driver location or (b) a driver credential value, and (iii) driver vehicle details associated with the plurality of drivers including at least one of: (a) a fuel level or (b) a fuel consumption rate;

extract, by executing the character recognition routine, the order data from the dispatch intake form, wherein both the pick-up yard and the destination location are railway terminals;

determine an order status corresponding to the order based on the order data, wherein the order status includes at least one of: (i) an arrived status indicating a train transporting the order has arrived at the pick-up yard, (ii) a placed status indicating the order is ready for offloading from the train, or (iii) a grounded status indicating the order is ready for transportation from the pick-up yard to the destination location;

retrieve, by accessing a plurality of dispatch applications (apps) of a plurality of drivers via an application programming interface (API), location information corresponding to a plurality of drivers;

analyze, by the dispatch learning model, (i) the order data in the universal format, (ii) the driver details, (iii) the driver vehicle details, and (iv) the location information to generate the driver logistics prediction value corresponding to the order data, wherein the dispatch learning model is trained or provided with the logistics data regarding order data of respective orders,

wherein the dispatch learning model is configured to output the one or more driver logistics prediction values corresponding to the order data of the respective orders, and wherein the driver logistics prediction value represents a time for a driver to complete the order based on (i) the driver location and (ii) time loss resulting from the driver refueling a vehicle corresponding to the driver vehicle details;

push the driver determination to a respective dispatch app of the driver, wherein, responsive to receiving the driver determination, the respective dispatch app displays the graphical user interface (GUI) depicting the driver determination; and

responsive to the driver accepting the order via the driver determination, receive the updated order status from the respective dispatch app of the driver indicating that the order is assigned to the driver.

36 . The tangible, non-transitory computer-readable medium of claim 30 , wherein the driver proceeds to deliver the order using the optimal route.

37 . The tangible, non-transitory computer-readable medium of claim 36 , wherein the instructions further cause the one or more processors to:

receive a second updated order status from the dispatch app of the driver indicating a completion of the order by the driver.

38 . The tangible, non-transitory computer-readable medium of claim 30 , wherein the driver delivers the order before a cut-off time to prevent a corresponding delay.

39 . The tangible, non-transitory computer-readable medium of claim 30 , wherein the instructions further cause the one or more processors to:

cause the dispatch app of the driver to display a second GUI including a first portion of the driver determination, the first portion comprising a scheduled order section that displays accepted orders of the driver on a particular date including the order,

track (i) an in-gate time of the driver arriving at the pick-up yard on the particular date and (ii) an outgate time of the driver departing the pick-up yard,

include (i) the in-gate time of the driver arriving at the pick-up yard on the particular date and (ii) the outgate time as entries in the historical record to maintain the historical record,

cause the dispatch app of the driver to display a third GUI including a second portion of the driver determination, the second portion comprising an order travel details section and a selectable navigation button, the order travel details section indicating the pick-up yard of the order and the destination location of the order, and

responsive to receiving a selection indication of the selectable navigation button from the driver, causing the dispatch app of the driver to render a graphical display of a second optimal route associated with the driver to provide the driver with travel instructions from the pick-up yard to the destination location.