Decision support system and method for establishing transportation plans

Abstract

The present invention relates to a system and method that supports the derivation of evidence-based results in establishing a traffic plan and making decisions in a city by collecting data such as traffic volume, public transportation usage rates, and public transportation usage patterns, and through simulation, analysis, and prediction. It includes: a data collection module that collects data from a database including transportation card usage data, credit card payment data, road network data, pedestrian traffic data, weather data, and accident data; a traffic simulation engine that predicts traffic flow for each candidate transfer center site based on data provided by the data collection module; a multi-criteria decision support module that analyzes multiple criteria including transportation accessibility, usage demand, transfer distance, cost, and environmental impact to evaluate candidate transfer center sites, and determines the optimal location for the transfer center by assigning weights to the multiple criteria; and a Geographic Information System (GIS)-based visualization module that displays the optimal location determined by the multi-criteria decision module on a map and visually represents traffic density.

Description

Technology Field

[0001] The present invention relates to a system and method that supports the derivation of evidence-based results in establishing a traffic plan and making decisions in a city by collecting data such as traffic volume, public transportation usage rates, and public transportation usage patterns, and through simulation, analysis, and prediction. Background Technology

[0002] Current urban transportation planning requires considering various variables such as complex data, population movement patterns, public transportation usage, road networks, and commercial facility density; however, existing methods tend to rely heavily on expert experience and limited statistical data. For instance, urban transportation planning tasks like selecting the location for a transfer center necessitate a comprehensive analysis of diverse traffic data, yet existing systems are limited to analyzing individual data and lack the capability to predict future traffic changes.

[0003] Recently, technologies supporting traffic planning in a more scientific and quantitative manner by utilizing big data, digital twins, artificial intelligence (AI), and multi-criteria decision-making (MCDM) techniques are being researched, but a sophisticated decision support system that incorporates real-time data has not yet been developed.

[0004] To implement a decision support system that predicts future traffic conditions and establishes optimal traffic policies and plans, technologies such as Big Data, Digital Twin, AI-based traffic simulation, GIS (Geographic Information System), and data standardization are utilized. Big Data-based traffic analysis technology analyzes traffic patterns using big data and processes real-time traffic data to formulate optimal traffic plans; it is applied to traffic data collection and management, as well as exploratory data analysis using big data. Digital Twin-based traffic simulation technology, which replicates real-world traffic infrastructure and environments in a virtual space, generates optimal scenarios for traffic planning. Technologies are being utilized to visualize and analyze road networks, traffic flow, and congestion areas using spatial data based on GIS, as well as various methods for proposing optimal policies and evaluating diverse scenarios based on the results of traffic data analysis. The problem to be solved

[0005] The present invention aims to develop a decision support system for establishing traffic plans, such as the optimal location selection for transfer centers. Existing methods have limitations in predicting future demand as they rely on historical traffic data; therefore, it is necessary to perform more sophisticated predictions by combining real-time traffic data with simulations.

[0006] Since existing systems analyze based on buses and subways, they fail to support the selection of optimal transfer center locations considering multiple transportation modes; therefore, this invention aims to perform multi-criteria analysis by reflecting complex factors such as public transportation, pedestrian movement, and vehicle flow.

[0007] For example, since there is a lack of a function to review in advance the effects of constructing a transfer center in a specific area, the present invention aims to present an optimal alternative by performing simulations on various candidate locations.

[0008] The problems solved by the present invention are not limited to those mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0009] As a means to solve the aforementioned technical problem, according to an embodiment of the present invention, the method comprises: a data collection module that collects data from a database including transportation card usage data, credit card payment data, road network data, floating population data, weather data, and accident data; a traffic simulation engine that predicts traffic flow for each candidate transfer center site based on data provided from the data collection module; a multi-criteria decision support module that analyzes multi-criteria factors including traffic accessibility, usage demand, transfer distance, cost, and environmental impact to evaluate candidate transfer center sites, and determines the optimal transfer center site by assigning weights to the multi-criteria factors; and a Geographic Information System (GIS)-based visualization module that displays the optimal site determined by the multi-criteria decision module on a map and visually represents traffic density.

[0010] It further includes a user interface module provided to allow the user to adjust the weights of specific evaluation criteria, and a multi-criteria decision module that performs recalculation based on the changed weights.

[0011] The above user interface module recalculates by reflecting the weights of specific evaluation criteria entered by the user in real time and provides an interface to compare and analyze simulation results.

[0012] The above data collection module collects data from local governments, public transportation operators, the Road Traffic Authority, mobile carriers, and the Korea Meteorological Administration, and includes a data preprocessing unit that normalizes and standardizes the data.

[0013] The above traffic simulation engine analyzes predicted traffic flow, including buses, subways, taxis, bicycles, and walking, on an hourly or segmented basis.

[0014] The above traffic simulation engine is configured to predict traffic flow based on changes in variables by applying scenarios such as traffic signal systems, road construction, and accident occurrences, and can recommend traffic plans by simultaneously comparing and analyzing multiple scenarios.

[0015] The above multi-criteria decision support module analyzes optimal candidate sites by including at least one factor among transportation accessibility, usage demand, transfer distance, cost, and environmental impact, and applies user input values ​​or predefined criteria values ​​as weights.

[0016] The above multi-criteria decision support module dynamically adjusts weights according to user preferences, applies machine learning algorithms to learn from the user's past decision data, and automatically recommends weights.

[0017] The above Geographic Information System (GIS)-based visualization module represents changes in traffic volume and pedestrian flow as a heatmap and can variably display data by time period or by section.

[0018] The above-mentioned Geographic Information System (GIS)-based visualization module is configured to display various data layer by layer according to user selection on a digital map, and can output changes in traffic volume over time in the form of a 3D model or animation.

[0019] The method comprises: a data collection step for making decisions to establish a transportation plan, including public transportation data such as transportation card transaction history, bus operation records, and bus route information; spatial data including road networks; and public data including information on total floor area by building use; a data preprocessing step for performing normalization and standardization of the data collected in the data collection step; a step of receiving parameter values ​​for conditions including construction costs, travel distance, and service levels from a user; a step of determining weights for the parameter values ​​for the conditions; a step of performing a simulation including traffic volume prediction, traffic congestion pattern analysis, and public usage pattern analysis using the collected data and the parameter values ​​entered by the user; a step of performing multi-criteria decision-making by applying the parameter values ​​for the conditions to the results of the simulation; and a step of providing the results of the multi-criteria decision-making to the user by combining them with a Geographic Information System (GIS). Effects of the invention

[0020] Since the present invention lacks the function to review in advance the effects of constructing a transfer center in a specific area by performing multi-criteria analysis reflecting complex factors such as public transportation, pedestrian movement, and vehicle flow, the present invention can present an optimal alternative by performing simulations on various candidate locations. Brief explanation of the drawing

[0021] FIG. 1 is a conceptual diagram of a decision support system according to the present invention. FIG. 2 is a configuration diagram of a decision support system according to the present invention. FIG. 3 is an example diagram of data collection types and configurations according to the present invention. FIG. 4 is an example of data constructed according to the present invention. FIG. 5 is an example diagram of data collection and preprocessing according to the present invention. FIG. 6 is an example of the main screen of the system according to the present invention. FIG. 7 is an example diagram showing a traffic volume layer displayed in 3D on a GIS according to the present invention. FIG. 8 is an example diagram showing a pedestrian traffic layer displayed in 2D on a GIS according to the present invention. FIG. 9 is an example diagram before applying the optimal transfer center location scenario on a GIS according to the present invention. FIG. 10 is an example diagram after applying the optimal transfer center site scenario on a GIS according to the present invention. Specific details for implementing the invention

[0022] Further objects, features, and advantages of the present invention can be more clearly understood from the following detailed description and the accompanying drawings.

[0023] Before providing a detailed description of the present invention, it should be understood that the present invention is capable of various modifications and may have various embodiments, and that the examples described below and illustrated in the drawings are not intended to limit the present invention to specific embodiments, but rather include all modifications, equivalents, and substitutions that fall within the spirit and scope of the present invention.

[0024] Hereinafter, specific technical details to be implemented in the present invention will be described in detail with reference to the attached drawings.

[0025] The present invention consists of a data collection module, a simulation engine, a multi-criteria decision support module, a Geographic Information System (GIS) visualization module, and a user interface (UI), and each module is interconnected to support accurate traffic pattern analysis and the establishment of optimal traffic plans.

[0026] The data collection module collects transportation card data, credit card usage history, road network data, and pedestrian traffic data in real time, and utilizes public data and IoT sensor data to secure information necessary for analyzing transfer center locations.

[0027] For example, the sources and contents for collecting transportation card usage data, credit card payment data, road network data, pedestrian traffic data, weather data, accident and event data, etc., are as shown in the following table.

[0028] Data name source explanation Transportation card usage data Local government, company Usage history for origin-destination analysis Credit card payment data credit card company Analysis of usage patterns of commercial facilities near the transfer center Road network data Korea Expressway Corporation, Ministry of Land, Infrastructure and Transport Road Congestion and Accessibility Analysis pedestrian traffic data Mobile carrier, sensor value Analysis of population movement flow by time of day Weather data Korea Meteorological Administration Prediction of changes in traffic flow due to weather Accident and event data National Police Agency, SNS Traffic impact assessment due to accidents and large-scale events

[0029] Since the raw data collected in this way may have different units or contain noise, duplicates, and missing values, a cleaning process is essential.

[0030] The cleaning method involves removing duplicate data and supplementing damaged data through data cleaning, normalization and standardization performed when units differ or deviations are large, and augmentation techniques applied when data is insufficient or unbalanced, such as rotation, scaling, and color jittering for image data, and synonym conversion and back translation for natural language data.

[0031] A digital twin-based traffic simulation engine performs the function of analyzing expected effects by simulating traffic flow for each candidate transfer center site in a three-dimensional virtual environment and predicting future traffic demand by applying an AI-based traffic pattern prediction algorithm. The present invention includes a simulation engine that predicts traffic flow for each candidate transfer center site by applying digital twin technology.

[0032] The functions of the above simulation engine include a traffic volume prediction model, traffic congestion pattern analysis, and public transportation usage pattern analysis.

[0033] The simulation engine described above performs scenario-based analysis involving multiple transportation modes (bus, subway, pedestrian, etc.). Specifically, it analyzes the movement patterns of individual users by applying Agent-Based Modeling techniques. Additionally, it predicts changes in traffic volume over time by applying cellular automata techniques to each candidate transfer center site. For instance, when performing the simulation, locations based on time and coordinates are defined as cells, and the analysis reflects the traffic volume and density of adjacent cells for those cells according to established criteria.

[0034] The Multi-Criteria Decision Making (MCDM) support module is a core technological element that enables optimal decisions to be made by considering complex factors during transportation planning. In this system, the MCDM module plays a role in performing various transportation-related decisions based on data, such as selecting transfer center locations, optimizing public transportation routes, and managing traffic flow. Areas where this multi-criteria decision-making support method can be applied include the optimal site selection for transfer centers, adjustment of bus and subway routes, optimization of road signaling systems, selection of locations for parking lots and shared mobility stations, establishment of alternative public transportation routes during disasters, and evaluation of transportation policies considering environmental impact.

[0035] The Multi-Criteria Decision Support Module selects the optimal location for a transfer center by considering factors such as traffic congestion, user convenience, construction costs, and environmental impact, and visually provides analysis results to enable policymakers to compare various scenarios. This system includes a Multi-Criteria Decision Making (MCDM) technique that evaluates various candidate sites to select the optimal location for the transfer center.

[0036] The multi-criteria decision support module collects traffic data from multiple sources, and the collected data undergoes normalization and standardization processes according to each decision criterion and variable. It also defines decision factors and derives optimal decisions based on them; for example, road network and public transportation accessibility can be defined as A, the number of public transportation users and pedestrian density as U, distance from existing modes of transportation as T, construction and operating costs as C, and environmental factors such as noise and carbon emissions as E.

[0037] Decision-making methods for selecting the optimal alternative by considering these multiple criteria include the Weighted Sum Model (WSM), which selects the optimal alternative by evaluating multiple criteria for each alternative, multiplying by weights for each criterion, and summing them; the Analytic Hierarchy Process, which assigns weights based on the preferences of experts or policymakers to hierarchically analyze the relative importance between decision criteria; the TOPSIS method, which selects the optimal alternative by measuring the distance from the solution and anti-ideal solution; and the ELECTRE method, which selects alternatives based on ranking after comparing superiority among alternatives by calculating the Concordance Index and Discordance Index and setting a threshold.

[0038] The multi-criteria decision support module can apply machine learning algorithms to learn from data on past user decisions and automatically recommend weights.

[0039] Machine learning is a process of learning patterns from data, performing predictions based on them, and building and improving models using empirical data without explicit programming. It provides a framework that uses mathematical models and algorithms to analyze data, identify patterns, and learn to predict future inputs. Machine learning is generally classified into three categories based on the learning method: supervised learning, unsupervised learning, and reinforcement learning. In this invention, it is preferable to apply a machine learning model based on a supervised learning method that learns from past decision data. Representative algorithms that can be applied include linear regression, polynomial regression, decision tree regression, and support vector regression (SVR).

[0040] After the decision criteria are determined by the methods described above, the policymaker can calculate a comprehensive score by applying weights to each criterion. As shown in the table below, the optimal candidate site score reflecting the weights for each candidate site can be calculated, and the optimal candidate site can be determined accordingly.

[0041] Evaluation items explanation weight Transportation accessibility (A) Road network and public transportation connectivity 0.3 Usage demand (U) Foot traffic and transportation card usage 0.25 Transfer distance (T) Distance traveled by other means of transportation 0.2 Cost (C) Construction costs and operating costs 0.15 Environmental impact (E) Environmental impact due to increased traffic volume 0.1

[0042] The Geographic Information System (GIS)-based visualization module visually represents traffic density, pedestrian flow, and public transportation routes on a map, and recommends optimal locations for transfer centers using heatmaps. It generates heatmaps that visually represent traffic volume, congestion, and pedestrian density, enabling comparative analysis of candidate sites. It also provides a user interface (UI) that allows for comparison of various scenarios and features that reflect optimal location results in real-time when policymakers adjust weights. The decision support interface allows policymakers to review optimal alternatives by adjusting various conditions such as cost, travel distance, and service level. Built as a web-based system, it is designed to be accessible from various devices.

[0043] FIG. 1 is a conceptual diagram of a system according to the present invention. The decision support system for establishing traffic plans represents a platform that collects population data, economic data, vehicle data, policy data, weather data, land use data, built environment data, and road data, and performs a simulation engine, a multi-criteria decision module, and visualization based on a geographic information system.

[0044] Figure 2 is a configuration diagram of a decision support system. The system consists of a data collection module, a simulation engine, a multi-criteria decision module, and a Geographic Information System (GIS)-based visualization module.

[0045] Figure 3 illustrates the process of data collection and processing. Public transportation information, such as daily transportation card transaction records, bus stop status by route, daily bus operation records, and bus route information, is collected; spatial information, such as road networks and grid data, is collected; and public data, such as information on total floor area by building use, is collected. The collected data is preprocessed through normalization and standardization to enable processing by the simulation engine.

[0046] FIG. 4 illustrates an example of data and a model constructed according to the present invention.

[0047] FIG. 5 is an exemplary diagram illustrating the data collection process according to the present invention. It shows an example in which bus route and stop data, building data, etc., are processed by a data collection module based on Busan City data constructed as a pilot project in the present invention.

[0048] FIG. 6 is an example screen of a decision support system developed to derive the optimal location for a new tram line layout and a transfer center within the Busan North Port Redevelopment Area as an embodiment according to the present invention. The purpose of this system is to select an optimal transfer center that reflects the expected influx of floating population after the completion of the North Port Redevelopment Area, considers the relationship with Busan Station near the area, pedestrian accessibility for various visitors including those with mobility impairments, and connectivity with nearby public transportation facilities such as Busan Station. The prototype screen features a map view with a parameter input panel on the left and a layer control window on the right. The parameter input panel allows setting network structure parameters for the departure and arrival stations and service types, coverage parameters for the walking distance within the range of major and auxiliary services, and the number of operating vehicles per hour to define and set high-frequency, medium-frequency, evening and weekend, and late-night service levels.

[0049] FIG. 7 is a screen showing traffic volume in 3D layers according to the present invention. It is designed so that areas with high traffic volume can be visually recognized.

[0050] FIG. 8 is a screen showing the flow of people according to the present invention as a 2D layer, and FIG. 9 is an example diagram showing the optimal transfer center overlaid before selecting it by reflecting traffic volume and flow of people.

[0051] FIG. 10 is an example screen diagram showing the optimal location of a transfer center finally displayed after multi-criteria decision-making according to the present invention.

Claims

Claim 1 A decision support system for establishing a traffic plan, comprising: a data collection module that collects data from a database including traffic card usage data, credit card payment data, road network data, floating population data, weather data, and accident data; a traffic simulation engine that predicts traffic flow for each candidate transfer center site based on data provided from the data collection module; a multi-criteria decision support module that analyzes multi-criteria factors including traffic accessibility, usage demand, transfer distance, cost, and environmental impact to evaluate candidate transfer center sites, and determines the optimal location for the transfer center by assigning weights to the multi-criteria factors; and a Geographic Information System (GIS)-based visualization module that displays the optimal location determined by the multi-criteria decision support module on a map and visually represents traffic density. Claim 2 A decision support system for establishing a traffic plan, further comprising: a user interface module provided to allow a user to adjust the weights of specific evaluation criteria, wherein a multi-criteria decision module performs recalculation according to the changed weights. Claim 3 A decision support system for establishing traffic plans according to paragraph 2, wherein the user interface module recalculates by reflecting the weight of a specific evaluation criterion entered by the user in real time and provides an interface to compare and analyze simulation results. Claim 4 In paragraph 1, the data collection module collects data from local governments, public transportation operating agencies, the Road Traffic Authority, mobile carriers, and the Korea Meteorological Administration, and includes a data preprocessing unit that normalizes and standardizes the data, making a decision support system for establishing traffic plans. Claim 5 A decision support system for establishing a traffic plan, wherein the traffic simulation engine in claim 1 is characterized by analyzing predicted traffic flow, including buses, subways, taxis, bicycles, and pedestrians, on an hourly or segmently basis. Claim 6 In claim 1, the traffic simulation engine is configured to predict traffic flow according to changes in variables by applying scenarios such as traffic signal systems, road construction, and accident occurrences, and is a decision support system for establishing traffic plans capable of recommending traffic plans by simultaneously comparing and analyzing multiple scenarios. Claim 7 In claim 1, the multi-criteria decision support module analyzes optimal candidate sites by including at least one element among traffic accessibility, usage demand, transfer distance, cost, and environmental impact, and the weights are determined by applying user input values ​​or predefined criteria values, in a decision support system for establishing a traffic plan. Claim 8 In claim 1, the multi-criteria decision support module dynamically adjusts weights according to user preferences and applies a machine learning algorithm to learn the user's past decision data and automatically recommends weights, making it a decision support system for traffic planning. Claim 9 In paragraph 1, the above-mentioned Geographic Information System (GIS)-based visualization module represents changes in traffic volume and floating population as a heatmap and is a decision support system for establishing traffic plans capable of variably displaying data by time period or by section. Claim 10 The decision support system for traffic planning according to claim 1, wherein the above-mentioned Geographic Information System (GIS)-based visualization module is configured to display various data layer by layer according to user selection on a digital map, and can output changes in traffic volume by time period in the form of a 3D model or animation. Claim 11 A decision-making method for establishing a transportation plan, comprising: a data collection step including public transportation data including transportation card transaction history, bus operation records, and bus route information, spatial data including road networks, and public data including information on total floor area by building use; a data preprocessing step for performing normalization and standardization of the data collected in the data collection step; a step of receiving parameter values ​​for conditions including construction costs, travel distance, and service levels from a user; a step of determining weights for the parameter values ​​for the conditions; a step of performing a simulation including traffic volume prediction, traffic congestion pattern analysis, and public usage pattern analysis using the collected data and the parameter values ​​input by the user; a step of performing multi-criteria decision-making by applying the parameter values ​​for the conditions to the results of the simulation; and a step of providing the results of the multi-criteria decision-making to the user by combining them with a Geographic Information System (GIS).

Images