Artificial intelligence-based logistics distribution path optimization and dynamic scheduling method and system

Abstract

The application discloses a logistics distribution path optimization and dynamic scheduling method and system based on artificial intelligence, relates to the technical field of logistics distribution, and comprises the following steps: collecting real-time data, processing the collected data, and obtaining a data set for path optimization and scheduling; constructing a logistics distribution path optimization model, calculating an optimal distribution path; and dynamically scheduling distribution vehicles and adjusting in real time according to real-time traffic and demand changes. The application adopts real-time feedback data to realize dynamic path optimization, continuously collects traffic, weather, vehicle state and cargo temperature and humidity data in the distribution process, and adjusts the path in real time. Various factors are comprehensively considered to improve the accuracy of path scoring and the distribution efficiency. Through calculation of the urgency of an order and the characteristics of goods, the dynamic scheduling system of the application can intelligently adjust the priority of the order, and is suitable for various environmental conditions.

Description

Technical Field

[0001] This invention relates to the field of logistics and distribution technology, and in particular to a method and system for optimizing and dynamically scheduling logistics and distribution routes based on artificial intelligence. Background Technology

[0002] In modern logistics and distribution, route optimization and dynamic scheduling are core issues for improving delivery efficiency and reducing costs. Current logistics management systems primarily rely on route planning based on static rules or simple traffic data scheduling. However, these methods are clearly insufficient in dynamically changing environments and struggle to meet complex logistics needs. These methods calculate routes before delivery begins and cannot dynamically adjust based on real-time traffic, weather changes, or vehicle status. Unexpected events (such as road construction or traffic accidents) can render existing routes invalid, failing to provide immediate optimization solutions. Traditional route optimization is largely based on fixed map data and traffic flow information, lacking consideration for factors such as weather, vehicle load, and special cargo requirements. Summary of the Invention

[0003] In view of the above-mentioned problems, the present invention is proposed.

[0004] Therefore, the technical problem solved by this invention is: how to dynamically adjust the path based on consideration of multiple influencing factors.

[0005] To address the aforementioned technical problems, this invention provides the following technical solution: collecting real-time data, processing the collected data to obtain a dataset for route optimization and scheduling; constructing a logistics delivery route optimization model to calculate the optimal delivery route; and dynamically scheduling delivery vehicles, making immediate adjustments based on real-time traffic and demand changes.

[0006] As a preferred embodiment of the AI-based logistics delivery route optimization and dynamic scheduling method described in this invention, the real-time data is collected through sensors, a logistics management system, and external data interfaces during the logistics delivery process, and the processing includes cleaning, noise reduction, missing value imputation, and data normalization.

[0007] As a preferred embodiment of the AI-based logistics delivery route optimization and dynamic scheduling method described in this invention, the logistics delivery route optimization model is expressed as follows:

[0008]

[0009] Where Z represents the total delivery cost, and N represents the number of delivery stations. To from the site Arrive at the station Transportation costs or distance, To from the site Arrive at the station Real-time traffic flow To from the site Arrive at the station The delivery time Choose decision variables for the path.

[0010] As a preferred embodiment of the AI-based logistics delivery route optimization and dynamic scheduling method described in this invention, the logistics delivery route optimization model is constrained, including vehicle capacity constraints, station access constraints, and time constraints.

[0011] The vehicle capacity constraint is expressed as follows: ;

[0012] The site access constraints are expressed as follows: ;

[0013] The time constraint is expressed as follows: ;

[0014] in, It is a website The demand, It is the maximum capacity of the vehicle. This is the maximum allowed time.

[0015] As a preferred embodiment of the AI-based logistics delivery route optimization and dynamic scheduling method described in this invention, the dynamic scheduling includes calculating the basic route cost, route environmental adaptability, and order influencing factors to obtain a final route score for dynamic scheduling and adjustment. The formula for the basic route cost is expressed as:

[0016]

[0017] in, For the site Arrive at the station Basic consumption; For the site Arrive at the station In time The distance below; For the site Arrive at the station The estimated time; Real-time traffic conditions: 0 indicates complete congestion, 1 indicates unobstructed traffic; The road complexity is indicated by 0, where 0 represents extremely severe conditions and 1 represents normal conditions.

[0018] As a preferred embodiment of the AI-based logistics delivery route optimization and dynamic scheduling method described in this invention, the route environment adaptability is expressed as:

[0019]

[0020] in, It is an environmental adaptability index; To account for weather conditions, 0 represents extremely severe weather, and 1 represents normal weather. Indicates the smoothness of road traffic; For vehicle load.

[0021] As a preferred embodiment of the AI-based logistics delivery route optimization and dynamic scheduling method described in this invention, the order influencing factors are represented as follows:

[0022]

[0023] in, Rate the impact of orders; The urgency level of the order is indicated by 0 for a regular order and 1 for an urgent order. For special cargo, the higher the value, the stricter the transportation requirements.

[0024] As a preferred embodiment of the AI-based logistics delivery route optimization and dynamic scheduling method described in this invention, the route score is expressed as:

[0025]

[0026] in, Score the final path; eliminate For paths with a value less than 0, set a minimum score threshold θ2 and a maximum score threshold θ1. If the path score is greater than θ1, the path is stable and is selected as the current optimal path for vehicle allocation.

[0027] like If ∈[θ2,θ1], it indicates that the score is moderate and fluctuates. In this case, we need to determine if optimization is possible: If If the value is less than 0.5, indicating poor weather or road conditions, an alternative route should be chosen; if If the value is greater than 0.7, meaning the order is urgent, then this path will be forcibly prioritized for scheduling;

[0028] like If the path score is less than θ2, dynamic scheduling is triggered, and a new path combination is recalculated.

[0029] As a preferred embodiment of the AI-based logistics delivery route optimization and dynamic scheduling method described in this invention, wherein: when a problem is detected during delivery... If the score continues to decline, the path score will be recalculated and the scheduling adjusted; when the vehicle is in motion... If the situation exceeds expectations, the system will recalculate the optimal scheduling scheme.

[0030] Record all actual driving routes and calculate the predicted route score. The error between the actual feedback score and the actual score is adjusted. When the error exceeds a threshold, the weights of weather, traffic, and order urgency in the path scoring model are adjusted.

[0031] Secondly, another objective of this invention is to provide an artificial intelligence-based logistics delivery route optimization and dynamic scheduling system, including a data acquisition and preprocessing module, a route optimization calculation module, and a dynamic scheduling decision module. The data acquisition and preprocessing module is used to collect and preprocess traffic, weather, order, and vehicle status data in real time. The route optimization calculation module is used to construct a logistics delivery route optimization model and calculate the optimal delivery route. The dynamic scheduling decision module is used to intelligently adjust vehicle scheduling, order allocation, and route selection based on real-time feedback data to ensure optimal execution of delivery tasks in a dynamic environment.

[0032] The beneficial effects of this invention are as follows: This invention provides an artificial intelligence-based logistics delivery route optimization and dynamic scheduling method and system. It utilizes real-time feedback data to achieve dynamic route optimization, continuously collecting traffic, weather, vehicle status, and cargo temperature and humidity data during the delivery process to adjust the route in real time. By comprehensively considering multiple factors, it improves the accuracy of route scoring and increases delivery efficiency. By calculating the urgency of orders and cargo characteristics, the dynamic scheduling system of this invention can intelligently adjust order priorities, adapting to various situations and environments. Attached Figure Description

[0033] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 The above is an overall flowchart of an artificial intelligence-based logistics delivery route optimization and dynamic scheduling method provided in one embodiment of the present invention. Detailed Implementation

[0035] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0036] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0037] Example 1

[0038] Reference Figure 1 As an embodiment of the present invention, a method for optimizing and dynamically scheduling logistics delivery routes based on artificial intelligence is provided, comprising:

[0039] S1: Collect real-time data, process the collected data, and obtain a dataset for path optimization and scheduling;

[0040] Furthermore, real-time data during the logistics and distribution process is collected through various sensors, logistics management systems, and external data interfaces, including but not limited to the location and speed of delivery vehicles, traffic conditions, weather conditions, customer needs, cargo information, and warehouse locations.

[0041] It should be noted that weather data includes data on weather changes such as temperature, precipitation, and wind speed.

[0042] Traffic data: Real-time traffic information, including traffic congestion, road construction, traffic accidents, etc.

[0043] Vehicle data: Basic information about the vehicle, including load capacity, fuel efficiency, and consumption.

[0044] Cargo data: cargo type and its specific requirements (such as cold chain requirements, weight, etc.).

[0045] Delivery time and route history data: Records past delivery routes and times for subsequent training.

[0046] Delivery target data: including the urgency level and order type of each delivery point.

[0047] The preprocessing process includes data cleaning, normalization, and feature extraction to ensure the quality and consistency of the data input to the model.

[0048] S2: Construct a logistics delivery route optimization model and calculate the optimal delivery route;

[0049] Furthermore, the logistics delivery route optimization model can be expressed as:

[0050]

[0051] (Total delivery cost): The optimization objective is to minimize the total delivery cost.

[0052] (Shipping cost or distance): From the delivery station Arrive at the station The actual transportation cost or distance is collected through maps and GPS.

[0053] (Traffic conditions): From the station Arrive at the station Real-time traffic flow Based on traffic monitoring data, 0 indicates severe traffic congestion, while 1 indicates unimpeded traffic.

[0054] (Transportation time): From station Arrive at the station The transportation time is calculated based on traffic flow and road condition information.

[0055] (Route selection decision variables): When choosing from station Arrive at the station hour, Otherwise, it is 0, indicating whether to select this path.

[0056] Constraints are imposed on the logistics delivery route optimization model, including vehicle capacity constraints, station access constraints, and time constraints;

[0057] Vehicle capacity constraints are expressed as follows: ;

[0058] Site access constraints are represented as follows: ;

[0059] The time constraint is expressed as: ;

[0060] in, It is a website To meet demand, each site must be visited once to ensure that every delivery site is served. It is the maximum capacity of the vehicle. This is the maximum allowed time.

[0061] S3: Dynamically dispatch delivery vehicles and make immediate adjustments based on real-time changes in traffic and demand.

[0062] Furthermore, dynamic scheduling involves calculating the basic path cost, path environmental adaptability, and order influencing factors separately to obtain the final path score for dynamic scheduling and adjustment. The formula for the basic path cost is expressed as:

[0063]

[0064] (Basic path cost): Path Basic consumption.

[0065] (Transportation distance): Station Arrive at the station In time The distance below.

[0066] (Transportation Time): Station Arrive at the station In time The estimated time below.

[0067] (Real-time traffic conditions): The range is 0, which represents complete congestion, and 1, which represents unobstructed flow.

[0068] (Road complexity): Value range, 0 represents extremely bad conditions, 1 represents normal conditions.

[0069] It should be noted that the estimated time represents the expected travel time of a vehicle from station i to station j at time τ, and can be obtained through the following method:

[0070] (1) Real-time GPS traffic data

[0071] By tracking the vehicle's route using GPS devices, the average travel time for the same route in the past can be calculated.

[0072] It dynamically adjusts based on historical data and current road conditions.

[0073] (2) Traffic flow prediction model

[0074] By combining historical traffic flow data and real-time traffic data, machine learning models (such as LSTM and regression models) are used to predict the current time. .

[0075] Input data includes:

[0076] Traffic flow in the same time period over the past n days .

[0077] The current vehicle density and average speed of the road section.

[0078] (3) Traffic Status API

[0079] Obtain real-time estimated time of arrival (ETA) through third-party transportation APIs (such as Google Maps, Baidu Maps, and Gaode API).

[0080] The information returned by the API includes:

[0081] Current average vehicle speed .

[0082] Current traffic conditions (smooth traffic, slow traffic, congested).

[0083] Calculation formula:

[0084] Furthermore, path environment adaptability can be expressed as:

[0085]

[0086] (Environmental Adaptability Index): Path The ability to adapt to weather and road conditions.

[0087] (Weather impact): Value range: 0 represents extremely severe weather, and 1 represents normal weather.

[0088] (Road accessibility): Considering complex road conditions, such as slope, construction, and road closures.

[0089] (Vehicle load): The real-time load of vehicles on this route.

[0090] It should be noted that road traffic flow It reflects the smoothness of a certain road, with a value range of [0,1], where 0 indicates complete congestion and 1 indicates unobstructed traffic.

[0091] (1) Traffic monitoring system

[0092] The city's traffic management system (cameras, geomagnetic sensors) collects data on vehicle flow and average speed for each road segment.

[0093] Calculate vehicle density:

[0094] in: It refers to the number of vehicles on the road segment. It is the total length of the road segment.

[0095] (2) Calculation of road traffic capacity

[0096] Calculate the vehicle throughput rate for this road segment within the current time period:

[0097] in: This is the maximum permissible vehicle density for that road segment (e.g., 100 vehicles per kilometer). A value close to 1 indicates smooth traffic, while a value close to 0 indicates severe congestion.

[0098] (3) In conjunction with traffic incidents

[0099] If the traffic API or cameras detect events such as road construction, car accidents, or abnormal weather, set... =0, forcing the road segment to become unavailable.

[0100] Furthermore, the factors influencing orders are expressed as follows:

[0101]

[0102] (Order Impact Score): The importance of orders to the route.

[0103] (Order urgency): Value range: 0 represents a regular order, and 1 represents an urgent order.

[0104] (Impact of special goods): The value range takes into account factors such as cold chain, fragile items, and timeliness. The larger the value, the stricter the transportation requirements.

[0105] It should be noted that special goods have an impact It reflects the external environmental impacts on a certain type of goods during the delivery process, mainly affecting perishable goods (cold chain), dangerous goods (chemicals), and fragile goods (glass products).

[0106] (1) Impact of cargo type

[0107] Obtain the goods type G through the order management system. k (such as fresh food and cold chain medicines).

[0108] If G k It requires a cold chain (such as low temperature requirements), and its special impact is significant.

[0109] (2) Environmental impact of transportation

[0110] Collect transportation environment data such as temperature, humidity, and vibration:

[0111] The vehicle-mounted temperature sensor records the current temperature of the cold chain warehouse. .

[0112] Vehicle humidity sensor records humidity .

[0113] Vibration sensors detect the degree of bumps. .

[0114] calculate:

[0115]

[0116] in: This is the optimal storage temperature for the goods. This is the optimal humidity level. This represents the maximum permissible vibration value for the cargo.

[0117] (3) Impact of goods validity period

[0118] If the goods have an expiration date, calculate the impact of the remaining delivery time:

[0119] in: This is the maximum permissible delivery time for goods. This represents the delivery time that has already elapsed.

[0120] Furthermore, the final path score is:

[0121] (Final Path Score): The overall score of the path; the higher the value, the better the path.

[0122] Eliminate For paths with a value less than 0, set a minimum score threshold θ2 and a maximum score threshold θ1. If the path score is greater than θ1, the path is stable and is selected as the current optimal path for vehicle allocation.

[0123] like If ∈[θ2,θ1], it indicates that the score is moderate and fluctuates. In this case, we need to determine if optimization is possible: If If the value is less than 0.5, indicating poor weather or road conditions, an alternative route should be chosen; if If the value is greater than 0.7, meaning the order is urgent, then this path will be forcibly prioritized for scheduling;

[0124] like If the path score is less than θ2, dynamic scheduling is triggered, and a new path combination is recalculated.

[0125] When detected during delivery If the score continues to decline, the path score will be recalculated and the scheduling adjusted; when the vehicle is in motion... If the situation exceeds expectations, the system will recalculate the optimal scheduling scheme.

[0126] Record all actual driving routes and calculate the predicted route score. The error between the actual feedback score and the actual score is adjusted. When the error exceeds a threshold, the weights of weather, traffic, and order urgency in the path scoring model are adjusted.

[0127] Example 2

[0128] As an embodiment of the present invention, an artificial intelligence-based logistics delivery route optimization and dynamic scheduling system is provided, including: a data acquisition and preprocessing module, a route optimization calculation module, and a dynamic scheduling decision module;

[0129] The data acquisition and preprocessing module is used to collect and preprocess traffic, weather, order, and vehicle status data in real time.

[0130] The route optimization calculation module is used to build a logistics delivery route optimization model and calculate the optimal delivery route;

[0131] The dynamic scheduling decision module is used to intelligently adjust vehicle scheduling, order allocation, and route selection based on real-time feedback data, ensuring that delivery tasks are executed optimally in a dynamic environment.

[0132] Example 3

[0133] In one embodiment of the present invention, in order to verify the beneficial effects of the present invention, a scientific demonstration is carried out through economic benefit calculation.

[0134] Experimental location: A large logistics park, including a warehousing center, multiple distribution stations, and urban and suburban roads.

[0135] Experimental vehicles: A total of 50 delivery vehicles, including 30 new energy vehicles and 20 traditional fuel vehicles.

[0136] Delivery orders: 500 orders will be randomly generated, covering general goods, cold chain goods, and large items, with delivery distances of 5-50km.

[0137] Data source:

[0138] GPS devices (to obtain vehicle location, speed, and dwell time).

[0139] Traffic monitoring system (to obtain real-time road conditions and congestion index).

[0140] Weather API (collects weather, temperature, humidity, and wind speed).

[0141] Order management system (records order urgency, goods type, and delivery time window).

[0142] Vehicle sensors (monitoring energy consumption, load, fuel status, and temperature of refrigerated trucks).

[0143] The experimental results are shown in the table below:

[0144]

[0145] The table shows the specific data on energy efficiency improvements during the experiment. Average delivery time decreased by 28.9% (traditional 45 minutes → optimized 32 minutes), delivery cost decreased by 29.2% (traditional 120 yuan → AI optimized 85 yuan), order on-time rate increased by 17%, route adjustment frequency decreased by 66.7%, and energy consumption decreased by 22.2%.

[0146] It significantly reduces delivery time, optimizes delivery costs, and improves order fulfillment rates, while also reducing energy consumption and enhancing the intelligence of logistics. Based on intelligent computing, real-time feedback, and adaptive optimization strategies, it can be widely applied to scenarios such as intelligent logistics, unmanned delivery, warehouse management, and intelligent cold chain distribution, thereby improving the operational efficiency of the logistics industry.

[0147] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for optimizing and dynamically scheduling logistics delivery routes based on artificial intelligence, characterized in that: include: Collect real-time data, process the collected data, and obtain a dataset for path optimization and scheduling; Construct a logistics delivery route optimization model and calculate the optimal delivery route; The logistics delivery route optimization model is expressed as follows: Where Z represents the total delivery cost, and N represents the number of delivery stations. To from the site Arrive at the station Transportation costs or distance, To from the site Arrive at the station Real-time traffic flow To from the site Arrive at the station The delivery time Choose decision variables for the path; The logistics delivery route optimization model is constrained, including vehicle capacity constraints, station access constraints, and time constraints. The vehicle capacity constraint is expressed as follows: ; The site access constraints are expressed as follows: ; The time constraint is expressed as follows: ; in, It is a website The demand, It is the maximum capacity of the vehicle. The maximum allowed time; The dynamic scheduling includes calculating the basic path cost, path environmental adaptability, and order influencing factors to obtain a final path score for dynamic scheduling and adjustment. The formula for the basic path cost is as follows: in, For the site Arrive at the station Basic consumption; For the site Arrive at the station In time The distance below; For the site Arrive at the station The estimated time; Real-time traffic conditions: 0 indicates complete congestion, 1 indicates unobstructed traffic; The road complexity is indicated by 0 for extremely severe conditions and 1 for normal conditions. The path environment adaptability is represented as follows: in, It is an environmental adaptability index; To account for weather conditions, 0 represents extremely severe weather, and 1 represents normal weather. Indicates the smoothness of road traffic; For vehicle load; The factors influencing orders are represented as follows: in, Rate the impact of orders; The urgency level of the order is indicated by 0 for a regular order and 1 for an urgent order. For special cargo, a higher value indicates stricter transportation requirements; Delivery vehicles are dynamically dispatched and adjusted in real time according to changes in traffic and demand.

2. The method for optimizing and dynamically scheduling logistics delivery routes based on artificial intelligence as described in claim 1 is characterized in that: The real-time data is collected through sensors, logistics management systems, and external data interfaces during the logistics and distribution process. The processing includes cleaning, noise reduction, missing value imputation, and data normalization.

3. The method for optimizing and dynamically scheduling logistics delivery routes based on artificial intelligence as described in claim 2 is characterized in that: The path score is expressed as: in, Score the final path; eliminate For paths with a value less than 0, set a minimum score threshold θ2 and a maximum score threshold θ1. If the path score is greater than θ1, the path is stable and is selected as the current optimal path for vehicle allocation. like If ∈[θ2,θ1], it indicates that the score is moderate and fluctuates. In this case, we need to determine if optimization is possible: If If the value is less than 0.5, indicating poor weather or road conditions, an alternative route should be chosen; if If the value is greater than 0.7, meaning the order is urgent, then this path will be forcibly prioritized for scheduling; like If the path score is less than θ2, dynamic scheduling is triggered, and a new path combination is recalculated.

4. The method for optimizing and dynamically scheduling logistics delivery routes based on artificial intelligence as described in claim 3 is characterized in that: When detected during delivery If the score continues to decline, the path score will be recalculated and the scheduling adjusted; when the vehicle is in motion... If the situation exceeds expectations, the system will recalculate the optimal scheduling scheme. Record all actual driving routes and calculate the predicted route score. The error between the actual feedback score and the actual score is adjusted. If the error exceeds a threshold, the weights of weather, traffic, and order urgency in the path scoring model are adjusted.

5. A system employing the artificial intelligence-based logistics distribution route optimization and dynamic scheduling method as described in any one of claims 1 to 4, characterized in that, include: Data acquisition and preprocessing module, path optimization calculation module, dynamic scheduling decision module; The data acquisition and preprocessing module is used to collect and preprocess traffic, weather, order, and vehicle status data in real time. The path optimization calculation module is used to construct a logistics delivery path optimization model and calculate the optimal delivery path. The dynamic scheduling decision module is used to intelligently adjust vehicle scheduling, order allocation, and route selection based on real-time feedback data, ensuring that delivery tasks are executed optimally in a dynamic environment.

Images