Harbor district trailer multi-target path optimization method and system based on improved Dijkstra
By constructing a directed graph of port area roads using an improved Dijkstra algorithm and combining it with a time-energy multi-objective cost function, the problems of high energy consumption and insufficient time constraints in port area trailer route planning were solved. This achieved time-energy optimal route optimization, improving the efficiency and feasibility of port area trailer scheduling.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANGHAI MARITIME UNIVERSITY
- Filing Date
- 2026-01-29
- Publication Date
- 2026-05-15
AI Technical Summary
Existing port area trailer route planning methods fail to take energy consumption factors into account, resulting in high overall energy consumption; some multi-objective route optimization methods lack effective constraints on transportation time, resulting in insufficient feasibility of optimization results in actual port area operations.
Based on the improved Dijkstra algorithm, a directed graph of port area roads is constructed. A multi-objective cost function is calculated by combining travel time and energy consumption. Through multi-objective path search with hard time constraints, the port area trailer routes are optimized to achieve time- and energy-optimal path planning.
This approach achieves the goals of reducing overall energy consumption, improving port area truck scheduling efficiency, and ensuring the engineering feasibility of the route plan while maintaining timely transportation.
Smart Images
Figure CN122048211A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of port area trailer route planning, and in particular to a port area trailer multi-objective route optimization method and system based on an improved Dijkstra's algorithm. Background Technology
[0002] With the continuous expansion of port operations and the increasing complexity of port transportation organization, route planning for port trucks during container transshipment has gradually become a crucial technical aspect affecting port operational efficiency and energy utilization. Existing technologies for port truck route planning primarily focus on minimizing distance or travel time as the main optimization objective. While these methods are relatively simple to calculate and easy to implement, they often overlook the energy consumption differences resulting from the complex road conditions within the port area.
[0003] In actual port environments, road conditions exhibit significant heterogeneity. Different road sections vary considerably in terms of driving speed, start-stop frequency, and traffic organization. Using only distance or time as the basis for route selection can easily lead to trailers frequently traversing roads with low speeds and high start-stop frequencies, resulting in overall high energy consumption and hindering the achievement of port energy-saving operation goals. Furthermore, while some existing multi-objective route optimization methods incorporate energy consumption indicators for comprehensive optimization, they lack effective constraints on transportation time, easily resulting in route solutions that fail to meet the actual production requirements of the port in terms of timeliness. This limits the feasibility and widespread application value of these methods in engineering projects.
[0004] The overall issue can be summarized into the following two questions: (1) Existing path planning methods take a single time or distance index as the optimization objective, and fail to take into account energy consumption factors, resulting in high overall energy consumption; (2) Some multi-objective path optimization methods lack effective constraints on transportation time, resulting in insufficient feasibility of the optimization results in actual port operations. Summary of the Invention
[0005] The purpose of this invention is to overcome the shortcomings of the existing technology by providing a multi-objective path optimization method and system for port trailers based on an improved Dijkstra's algorithm, so as to solve or partially solve the problem that the existing path optimization methods fail to take into account energy consumption factors, resulting in high overall energy consumption.
[0006] The objective of this invention can be achieved through the following technical solutions: One aspect of the present invention provides a multi-objective path optimization method for port trailers based on an improved Dijkstra's algorithm, comprising the following steps: Obtain port area road network data, as well as the starting and destination nodes of trailer transportation tasks; Based on the port area road network data, a directed graph of port area roads is constructed; Obtain the trailer's travel time and energy consumption on various roads; Based on the directed graph of the port area roads, the baseline path of the trailer from the starting node of the transportation task to the target node is calculated using the Dijkstra path search algorithm with the shortest baseline time. By utilizing a pre-constructed multi-objective cost function based on the driving time and driving energy consumption, and through a multi-objective path search with hard time constraints, the final path with optimal time and energy consumption is obtained, thereby achieving multi-objective path optimization for port area trailers.
[0007] As a preferred technical solution, the calculation process for driving time and driving energy consumption includes the following steps: The travel time is calculated based on the road distance and the trailer's speed. Driving energy consumption is calculated based on road distance, travel time, and number of starts and stops.
[0008] As a preferred technical solution, the process of obtaining the baseline path using the Dijkstra path search algorithm with the shortest baseline time includes the following steps: Initialize a first container, dist, to record the shortest time from the starting node of the transportation task to each node, and a second container, prev, to record the predecessor node of each node on the shortest path. Initialize the value of the transportation task start node in the first container dist; Create a priority queue by constructing a tuple based on the cumulative time from the starting point to the current node and the node itself, and then adding it to the priority queue; In response to the priority queue being non-empty, the node u with the minimum cumulative time in the queue is retrieved. If the retrieved node u is the target node, the search ends. Otherwise, all adjacent nodes v of the retrieved node u are traversed, and the single-trip time time_uv from the retrieved node u to the adjacent node v is calculated. The shortest time from the starting node of the transportation task to u + the single-trip time time_uv from u to v is compared with the currently known shortest time from the starting point to v. If the former is smaller, the latest shortest time to v in the first container dsit is updated, and the optimal path to v recorded in the second container prev is from u. The new tuple is added to the priority queue, and this step is repeated. Starting from the target node, the previous node is found based on the second container, and the baseline path is obtained through recursion.
[0009] As a preferred technical solution, the multi-objective cost function is: in, , , Let be the coefficient, and , The normalized road distance. The normalized travel time This is the energy consumption calculated based on the number of start-stop cycles and then normalized.
[0010] As a preferred technical solution, the process of multi-objective path search with time hard constraints includes the following steps: During path expansion, the cumulative time of the current path is recorded. ; like If the current path is not extended, skip the current path; otherwise, perform a comprehensive cost update and add the candidate node to the priority queue. The preset maximum time, This refers to the travel time.
[0011] As the preferred technical solution, energy consumption is calculated using the following formula. in, , , For coefficients, , These are the road distance and travel time, respectively.
[0012] As a preferred technical solution, the following steps are also included: The optimal path is backtracked to obtain the complete trailer travel path. The total distance, total travel time and total energy consumption of the trailer travel path are statistically analyzed and compared to evaluate the effectiveness of the path optimization.
[0013] As a preferred technical solution, the process of constructing a directed graph of port area roads includes the following steps: Based on the port area road network data, the port area roads are modeled and abstracted into a graph structure consisting of nodes and directed edges. The road attributes are then standardized to form a directed graph including a set of nodes and a set of roads.
[0014] As a preferred technical solution, the port area road network data includes node information and road information within the port area, wherein the node information is the location within the port area, and the road information is the traffic relationship between nodes.
[0015] Another aspect of the present invention provides a port area trailer multi-objective path optimization system based on an improved Dijkstra's algorithm, used to implement the aforementioned port area trailer multi-objective path optimization method, the system comprising: The port area road network and task input module is used to acquire port area road network data, as well as the starting and target nodes of the truck's transportation task; The road network modeling and data preprocessing module is used to construct a directed graph of port area roads based on the port area road network data; The time and energy consumption module is used to obtain the trailer's driving time and driving energy consumption on various roads; The baseline path calculation module is used to calculate the baseline path of the trailer from the starting node of the transportation task to the target node based on the directed graph of the port area roads and using the Dijkstra path search algorithm with the shortest baseline time. The multi-objective path optimization module is used to obtain the time- and energy-optimized final path by using a pre-constructed multi-objective cost function based on the driving time and driving energy consumption, through time-constrained multi-objective path search, thereby realizing multi-objective path optimization for port area trailers.
[0016] Compared with the prior art, the present invention has at least one of the following beneficial effects: (1) Achieving port area trailer route optimization that takes into account both time and energy consumption: This invention first uses the directed graph of port area roads and the Dijkstra path search algorithm with the shortest baseline time to calculate the baseline path of the trailer from the starting node of the transportation task to the target node. Then, it uses a pre-constructed multi-objective cost function based on driving time and driving energy consumption to obtain the final path with the best time and energy consumption, thus achieving multi-objective route optimization for port area trailers. Through the unified multi-objective cost function, the comprehensive optimization of distance, time and energy consumption is achieved, reducing overall energy consumption and improving the efficiency of port area trailer scheduling while ensuring transportation timeliness.
[0017] (2) High feasibility: This invention obtains the time- and energy-optimal final path through multi-objective path search with hard time constraints. In the process of multi-objective path search, a hard time constraint mechanism is introduced to ensure the engineering feasibility of the path scheme. The parameters are configurable and applicable to different port area transportation scenarios. Attached Figure Description
[0018] Figure 1 This is a flowchart of the port area trailer multi-objective path optimization method based on the improved Dijkstra method in the embodiment; Figure 2 This is a schematic diagram of the port area trailer multi-objective path optimization system in the embodiment. Detailed Implementation
[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. 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 scope of protection of the present invention.
[0020] Example 1 To address the problems existing in the prior art, this embodiment provides a multi-objective path optimization method for port area trailers based on an improved Dijkstra's algorithm. (See [link to relevant documentation]). Figure 1 Taking the road network environment of a certain port area as an example, the steps include: Step S1: Port Area Road Network and Task Input. In this embodiment, port area road network data and trailer transportation task information are first acquired. The port area road network data includes node information and road information within the port area. Nodes represent key locations within the port area, and roads represent the traffic relationships between nodes. The transportation task information includes the starting node and target node of the trailer. For example, the starting node of the trailer is set as node A, and the target node is node F.
[0021] Step S2, Road Network Modeling and Data Preprocessing. Based on the port area road network data obtained in Step S1, the port area roads are modeled and processed, abstracting them into a graph structure composed of nodes and directed edges. The road attributes are standardized, and a port area road network model containing a set of nodes and a set of roads is constructed, providing a unified data structure foundation for subsequent path calculation.
[0022] Step S3: Calculation of Time and Energy Consumption Indicators. Based on the port area road network model, the travel time and energy consumption indicators of the trailer on each road are calculated according to the attribute information such as distance, speed, and number of starts and stops. The travel time is calculated from the road distance and speed, while the energy consumption indicator is determined by comprehensively considering factors such as road distance, travel time, and number of starts and stops. For example, in this embodiment, the distance from road A to B is 500 meters, and the speed is 10 m / s. Therefore, the travel time on this road is 50 seconds, and the corresponding energy consumption value E1 is calculated.
[0023] Step S4, Baseline Path Calculation. On the port area road network model, using road travel time as the sole weight, a shortest path search algorithm is employed to calculate the shortest time path from starting node A to target node F. The travel time of this shortest path is then used as the time benchmark for subsequent multi-objective path optimization. For example, in this embodiment, the calculated shortest time path is A→C→F, with a total travel time of 300 seconds.
[0024] Specifically, the search algorithm includes the following process: (1) Initialize basic data.
[0025] First, prepare two core state recording containers for all nodes in the graph.
[0026] `dist[node]` is used to record the shortest time from the starting point to the node. Initially, the `dist` value of all nodes is set to infinity, indicating that the path time to these nodes is unknown at the beginning.
[0027] `prev[node]` is used to record the predecessor node of the node on the shortest path, that is, which node to come from to get the shortest time. Initially, the `prev` value of all nodes is set to null, indicating that no predecessor relationship has been found yet.
[0028] (2) Set the initial state of the starting point.
[0029] If the dist value of the starting point is set to 0, the time taken from the starting point to itself will necessarily be 0.
[0030] (3) Initialize the priority queue.
[0031] Create a priority queue Q whose core feature is that it can quickly retrieve the element with the smallest cumulative time and add the tuple (0, start) to the queue. The first value is the cumulative time from the starting point to the node, and the second value is the node itself, which means that the cumulative time of the starting point is 0.
[0032] (4) Process the nodes in the queue in a loop.
[0033] As long as there are still elements in priority queue Q, repeat the following operation: 1. Take out the node u with the smallest cumulative time in the queue. The priority queue will be automatically sorted to ensure that each node u taken out is the one that has the shortest time from the starting point to the node.
[0034] 2. Check the termination condition. If the extracted node u is exactly the target node, end the search algorithm directly. At this point, the shortest time path to the target node has been found.
[0035] 3. Traverse all neighboring nodes v of node u. For each node v directly connected to u, perform the following calculations and judgments: Calculate the time taken for a single trip from u to v. Use the formula time_uv=distance(u,v) / speed(u,v) to calculate the time required to walk from u to v.
[0036] Relaxation operation. Compare the shortest time from the starting point to u, the single time from u to v dist[u]+time_uv, and the currently known shortest time from the starting point to v dist[v].
[0037] If the former is smaller, it means a better path to v has been found, then execute: Update dist[v] by changing it to dist[u]+time_uv, which records the latest and shortest time up to v.
[0038] Update prev[v] by changing prev[v] to u, and record that the optimal path to v is from u.
[0039] Enqueue update: Add the new tuple (dist[v],v) to the priority queue Q for use when processing other nodes later.
[0040] (5) Backtrack to obtain the complete shortest path.
[0041] After the search is complete, the path is deduced in reverse using the prev array: Starting from the target node, find its predecessor node by using prev[target node]; Then find the previous node through prev[previous node], and so on, until you backtrack to the starting point; Finally, reversing the node sequence obtained by backtracking gives the complete shortest time path.
[0042] Step S5, Multi-objective Path Optimization. Under the time baseline constraint obtained in Step S4, a multi-objective weighted cost function including path distance, travel time, and energy consumption indicators is constructed. A hard time constraint is introduced during the path search process to expand and filter candidate paths. When the cumulative travel time of a candidate path exceeds the threshold allowed by the time baseline, the candidate path is discarded. When a candidate path meets the time constraint, its comprehensive cost is calculated and compared with the path optimization, thereby obtaining the path scheme that satisfies the time constraint and has the optimal comprehensive cost. For example, in this embodiment, the final optimal path is A→B→D→F, with a travel time of 320 seconds and a total energy consumption lower than that of the shortest time path.
[0043] Specifically, the logic for constructing the multi-objective cost function is as follows: For each road edge e, calculate: = Road distance; = / Driving speed; ; Among them, the energy consumption model parameters a, b, and c are positive real numbers, used to adapt to different vehicle models and port operating conditions.
[0044] To each , , Perform normalization processing; The comprehensive cost function is defined as: in .
[0045] The specific logic for multi-objective path search with hard time constraints is as follows: During path expansion, the cumulative time of the current path is recorded. ; like If so, skip the roadside and do not extend the path further; Otherwise, perform a comprehensive cost update and add the candidate node to the priority queue.
[0046] The logic for path backtracking and result output is as follows: Starting from the target node, backtrack to the starting node level by level according to the prev pointer to obtain the optimal path; The system accumulates the distance, time, and energy consumption along each roadside in the path, and outputs the path results and performance indicators.
[0047] Step S6, Route Evaluation and Result Output. The optimal route obtained in Step S5 is backtracked to obtain the complete trailer driving route. The total distance, total driving time, and total energy consumption corresponding to this route are statistically analyzed. The evaluation results are output for analysis and comparison of the route optimization effect, completing one port area trailer route optimization process.
[0048] Example 2 Based on Example 1, this example provides a port area trailer multi-objective path optimization system based on an improved Dijkstra's algorithm. See [link to example]. Figure 2 The system includes: (1) Port area road network and task input module.
[0049] This system is used to receive port area road network data and trailer transportation task information. The port area road network data includes node information and road information within the port area; nodes represent key locations within the port area, and roads represent the traffic relationships between nodes. The transportation task information includes the starting and destination locations of the trailers.
[0050] (2) Road network modeling and data preprocessing module.
[0051] This process is used to process the input port area road network data, abstracting the port area roads into a directed graph structure. During this process, road attributes are standardized, and a port area road network model containing nodes and directed edges is constructed, providing a unified data structure foundation for subsequent path calculations.
[0052] (3) Time and energy consumption module.
[0053] This tool is used to calculate the trailer's travel time and energy consumption on each road based on the attribute information of each road in the road network model. The travel time is calculated based on road distance and travel speed, while the energy consumption takes into account factors such as road distance, travel time, and the number of starts and stops.
[0054] (4) Baseline path calculation module.
[0055] This module is used to perform baseline path search processing on the road network model. It employs a shortest path search algorithm with travel time as the sole weight to calculate the shortest time path from the starting node to the target node, and uses the travel time of this path as the time benchmark for subsequent multi-objective path optimization.
[0056] (5) Multi-objective path optimization module.
[0057] This module is used to perform multi-objective path optimization based on baseline path calculation results. It constructs a multi-objective weighted cost function that includes distance, time, and energy consumption, applies time constraints during the path search process, and obtains the optimal path scheme that satisfies the time constraints and has the best overall cost through an improved path search algorithm.
[0058] (6) Path evaluation and result output module.
[0059] This tool is used to evaluate the path results output by the multi-objective path optimization module, calculate the total distance, total travel time, and total energy consumption of the path, and output the evaluation results for analysis and comparison of the path optimization effect.
[0060] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A multi-objective path optimization method for port area trailers based on an improved Dijkstra's algorithm, characterized in that, Includes the following steps: Obtain port area road network data, as well as the starting and destination nodes of trailer transportation tasks; Based on the port area road network data, a directed graph of port area roads is constructed; Obtain the trailer's travel time and energy consumption on various roads; Based on the directed graph of the port area roads, the baseline path of the trailer from the starting node of the transportation task to the target node is calculated using the Dijkstra path search algorithm with the shortest baseline time. By utilizing a pre-constructed multi-objective cost function based on the driving time and driving energy consumption, and through a multi-objective path search with hard time constraints, the final path with optimal time and energy consumption is obtained, thereby achieving multi-objective path optimization for port area trailers.
2. The port area trailer multi-objective path optimization method based on improved Dijkstra's algorithm as described in claim 1, characterized in that, The calculation process for driving time and driving energy consumption includes the following steps: The travel time is calculated based on the road distance and the trailer's speed. Driving energy consumption is calculated based on road distance, travel time, and number of starts and stops.
3. The port area trailer multi-objective path optimization method based on improved Dijkstra's algorithm as described in claim 1, characterized in that, The process of obtaining the baseline path using the shortest-time Dijkstra's path search algorithm includes the following steps: Initialize a first container, dist, to record the shortest time from the starting node of the transportation task to each node, and a second container, prev, to record the predecessor node of each node on the shortest path. Initialize the value of the transportation task start node in the first container dist; Create a priority queue by constructing a tuple based on the cumulative time from the starting point to the current node and the node itself, and then adding it to the priority queue; In response to the priority queue being non-empty, the node u with the minimum cumulative time in the queue is retrieved. If the retrieved node u is the target node, the search ends. Otherwise, all adjacent nodes v of the retrieved node u are traversed, and the single-trip time time_uv from the retrieved node u to the adjacent node v is calculated. The shortest time from the starting node of the transportation task to u + the single-trip time time_uv from u to v is compared with the currently known shortest time from the starting point to v. If the former is smaller, the latest shortest time to v in the first container dsit is updated, and the optimal path to v recorded in the second container prev is from u. The new tuple is added to the priority queue, and this step is repeated. Starting from the target node, the previous node is found based on the second container, and the baseline path is obtained through recursion.
4. The port area trailer multi-objective path optimization method based on improved Dijkstra's algorithm as described in claim 1, characterized in that, The multi-objective cost function is: in, , , The coefficient is , and , The normalized road distance. The normalized travel time This is the energy consumption calculated based on the number of start-stop cycles and then normalized.
5. The port area trailer multi-objective path optimization method based on improved Dijkstra's algorithm as described in claim 4, characterized in that, The time-constrained multi-objective path search process includes the following steps: During path expansion, the cumulative time of the current path is recorded. ; like If the current path is not extended, skip the current path; otherwise, perform a comprehensive cost update and add the candidate node to the priority queue. The preset maximum time, This refers to the travel time.
6. The port area trailer multi-objective path optimization method based on improved Dijkstra's algorithm according to claim 4, characterized in that, Energy consumption is calculated using the following formula. in, , , For coefficients, , These are the road distance and travel time, respectively.
7. The port area trailer multi-objective path optimization method based on improved Dijkstra's algorithm as described in claim 1, characterized in that, It also includes the following steps: The optimal path is backtracked to obtain the complete trailer travel path. The total distance, total travel time and total energy consumption of the trailer travel path are statistically analyzed and compared to evaluate the effectiveness of the path optimization.
8. The port area trailer multi-objective path optimization method based on improved Dijkstra's algorithm according to claim 1, characterized in that, The process of constructing a directed graph of port area roads includes the following steps: Based on the port area road network data, the port area roads are modeled and abstracted into a graph structure consisting of nodes and directed edges. The road attributes are then standardized to form a directed graph including a set of nodes and a set of roads.
9. The port area trailer multi-objective path optimization method based on improved Dijkstra's algorithm according to claim 1, characterized in that, The port area road network data includes node information and road information within the port area, wherein the node information is the location within the port area, and the road information is the traffic relationship between nodes.
10. A port area trailer multi-objective path optimization system based on improved Dijkstra's algorithm, characterized in that, The system for implementing the port area trailer multi-objective path optimization method as described in any one of claims 1-9 includes: The port area road network and task input module is used to acquire port area road network data, as well as the starting and target nodes of the truck's transportation task; The road network modeling and data preprocessing module is used to construct a directed graph of port area roads based on the port area road network data; The time and energy consumption module is used to obtain the trailer's driving time and driving energy consumption on various roads; The baseline path calculation module is used to calculate the baseline path of the trailer from the starting node of the transportation task to the target node based on the directed graph of the port area roads and using the Dijkstra path search algorithm with the shortest baseline time. The multi-objective path optimization module is used to obtain the time- and energy-optimized final path by using a pre-constructed multi-objective cost function based on the driving time and driving energy consumption, through time-constrained multi-objective path search, thereby realizing multi-objective path optimization for port area trailers.