Red zone generation system and method for operating DRT platform

Abstract

The present invention relates to a red zone generation system and method for operating a DRT platform, wherein a red zone is generated using call information of a DRT vehicle, and a DRT call is restricted in the red zone. Accordingly, calls between red zone stops are restricted so that mobility rights of citizens in transport-disadvantaged areas can be secured.

Description

Red zone creation system and method for operating a DRT platform

[0001] The present invention relates to a red zone creation system and method for operating a DRT platform, and more particularly, to a red zone creation system and method for operating a DRT platform that creates a red zone using call information of a DRT vehicle and thereby restricts DRT calls within the red zone.

[0002] Demand Responsive Transport (DRT) is a type of public transportation that provides flexible service with flexible timing and location based on demand. It plays a crucial role, particularly in areas with limited public transportation, by adjusting routes and times to meet user demand. Traditional public transportation systems operate on fixed routes and schedules, making them inefficient in areas with low or fluctuating demand. DRT was introduced to overcome this limitation.

[0003] Most domestic DRTs are operated with the goal of guaranteeing the right to mobility for citizens in transportation-vulnerable areas. However, problems that arise during actual operation are preventing citizens in transportation-vulnerable areas from using them.

[0004] Specifically, because the current DRT system is based on call demand, it is experiencing excessive calls in densely populated, transit-disadvantaged areas with active public transportation. While transit-disadvantaged areas have ample alternative public transportation options, their high population densities lead to a concentration of DRT calls. This phenomenon, known as the "swamp effect," prevents DRT vehicles from leaving high-demand areas. This situation leads to dispatches being concentrated in transit-disadvantaged areas, resulting in longer wait times and call rejections for residents in transit-disadvantaged areas.

[0005] Therefore, in order to solve the limitations of the prior art described above, there is a need for a red zone creation system and method for operating a DRT platform that creates a red zone using call information of a DRT vehicle and thereby restricts DRT calls within the red zone.

[0006] The technical problem to be achieved by the present invention relates to a red zone creation system and method for operating a DRT platform that creates a red zone using call information of a DRT vehicle to guarantee the right to movement of citizens in traffic-vulnerable areas and thereby restricts DRT calls within the red zone.

[0007] Another technical task to be achieved by the present invention is a red zone creation system and method for operating a DRT platform that can select a minimum red zone according to a mathematical formula to minimize service restrictions for citizens in transportation-vulnerable areas.

[0008] In addition, the technical problem to be achieved by the present invention relates to a red zone creation system and method for operating a DRT platform that can variably select a red zone according to environmental conditions and the judgment of an operating entity in order to efficiently operate a DRT vehicle.

[0009] The technical problems to be solved by the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned can be clearly understood by a person having ordinary skill in the technical field to which the present invention belongs from the description below.

[0010] A red zone generation system for operating a DRT platform according to one embodiment of the present invention includes: a DRT demand data collection unit for collecting in real time DRT demand data including at least one of a number of DRT calls, call success and failure information, a call route, an operation route, identification information for each stop, and location information; a spatial clustering unit for forming spatial clusters by clustering stops in an analysis area based on the DRT demand data and spatial information; a spatial cluster visit count calculation unit for calculating the number of times each stop has visited each cluster; an objective function setting unit for setting an objective function based on the number of spatial cluster visits; a red zone representative stop extraction unit for extracting a red zone representative stop based on the degree of contribution to minimizing the objective function; a red zone additional stop extraction unit for selecting an additional stop based on a similarity determination result based on the extracted red zone representative stop; and a red zone reflection unit for setting a red zone area based on the extracted red zone representative stop and the red zone additional stop and reflecting the red zone area on a DRT map.

[0011] Here, minimizing the objective function means minimizing the variance of the number of visits to each cluster by each stop.

[0012] Here, the objective function setting unit can set an objective function that calculates a variance value by adding the squares of the differences between the number of visits to each cluster and the average number of visits.

[0013] In addition, the red zone representative stop extraction unit calculates a change in the objective function when the first stop is included in the red zone, selects a stop that minimizes this change, and, if additional calculation is required based on predetermined conditions, includes the first stop as an additional red zone stop and performs additional calculation.

[0014] Additionally, the above predetermined conditions can be determined using ModBIC (Modified Bayesian Information Criterion).

[0015] In addition, the red zone additional stop extraction unit may further include a determination based on an evaluation of OD (Origin-Destination) Pattern similarity or cost for distance when determining similarity between stops.

[0016] Additionally, the red zone reflection unit may include setting the DRT platform to restrict DRT calls within the set red zone and allow DRT calls to stops outside the red zone.

[0017] A method for generating a red zone by a system for generating a red zone for operating a DRT platform according to one embodiment of the present invention comprises the steps of: collecting DRT demand data in real time, including at least one of a number of DRT calls, call success and failure information, a call route, an operation route, identification information for each stop, and location information; forming spatial clusters by clustering stops in an analysis area based on the DRT demand data and spatial information; calculating the number of times each stop has visited each cluster; setting an objective function based on the number of spatial cluster visits; extracting a representative red zone stop based on the degree of contribution to minimizing the objective function; selecting an additional stop based on a similarity determination result based on the extracted representative red zone stop; and setting a red zone area based on the extracted representative red zone stop and the additional red zone stop and reflecting the red zone area on a DRT map.

[0018] Here, minimizing the objective function may be minimizing the variance of the number of visits to each cluster by each stop.

[0019] Here, the objective function setting unit can set an objective function that calculates a variance value by adding the squares of the differences between the number of visits to each cluster and the average number of visits.

[0020] In addition, the red zone representative stop extraction unit calculates a change in the objective function when the first stop is included in the red zone, selects a stop that minimizes this change, and, if additional calculation is required based on predetermined conditions, includes the first stop as an additional red zone stop to perform additional calculation.

[0021] Additionally, the above predetermined conditions can be determined using ModBIC (Modified Bayesian Information Criterion).

[0022] In addition, the red zone additional stop extraction unit may determine similarity between stops based on an evaluation of OD (Origin-Destination) Pattern similarity or cost for distance.

[0023] The above red zone reflection unit can set the DRT platform to restrict DRT calls within the set red zone and allow DRT calls to stops outside the red zone.

[0024] According to an embodiment of the present invention, red zones can be created to limit the operating area of ​​DRT vehicles. This can guarantee the right to mobility for citizens in vulnerable transportation areas by limiting calls between red zone stops.

[0025] Furthermore, according to an embodiment of the present invention, representative red zone stops and additional red zone stops can be extracted based on mathematical formulas. This allows for the selection of a minimum number of red zones, thereby minimizing service restrictions for citizens in transportation-disadvantaged areas.

[0026] Furthermore, according to embodiments of the present invention, variable settings can be changed based on environmental conditions and the operator's judgment. This allows for efficient operation of the DRT system based on environmental conditions by variably selecting the red zone.

[0027] The effects of the present invention are not limited to the effects described above, and should be understood to include all effects that can be inferred from the detailed description of the present invention or the composition of the invention described in the claims.

[0028] Figure 1 is a reference diagram illustrating an OD (Origin-Destination) pattern of a conventional DRT system to explain the necessity of setting a red zone.

[0029] FIG. 2 is a schematic diagram schematically illustrating a red zone creation system for operating a DRT platform according to an embodiment of the present invention.

[0030] Figure 3 is a block diagram of a red zone generation system according to an embodiment of the present invention.

[0031] FIG. 4 is a reference diagram illustrating a DRT operation route in which a red zone is set in a red zone generation system for operating a DRT platform according to an embodiment of the present invention.

[0032] FIG. 5 is a flowchart illustrating a red zone creation process for operating a DRT platform according to an embodiment of the present invention.

[0033] FIG. 6 is a flowchart illustrating a process of setting an objective function and extracting a red zone representative stop in a red zone creation method for operating a DRT platform according to an embodiment of the present invention.

[0034] FIG. 7 is a flowchart illustrating a process of extracting additional red zone stops and determining a red zone area in a red zone creation method for operating a DRT platform according to an embodiment of the present invention.

[0035] FIG. 8 is a reference diagram illustrating a method for evaluating distance similarity in a red zone creation method for operating a DRT platform according to an embodiment of the present invention.

[0036] FIG. 9 is a reference diagram illustrating a method for evaluating OD pattern similarity in a red zone creation method for operating a DRT platform according to an embodiment of the present invention.

[0037] Hereinafter, the present invention will be described with reference to the attached drawings. However, the present invention can be implemented in various different forms and is therefore not limited to the embodiments described herein. In the drawings, irrelevant parts have been omitted for clarity of description, and similar parts have been designated with similar reference numerals throughout the specification.

[0038] Throughout the specification, when a part is said to be "connected (connected, contacted, or coupled)" to another part, this includes not only cases where it is "directly connected," but also cases where it is "indirectly connected" with another part in between. Furthermore, when a part is said to "include" a component, this does not exclude other components, but rather implies that it may include other components, unless otherwise specifically stated.

[0039] The terminology used herein is merely used to describe specific embodiments and is not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this specification, it should be understood that the terms "comprises" or "has" indicate the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, but do not exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0040] The "Red Zone," as used in the embodiments of the present invention, refers to a specific area designated to maximize the efficiency of a Demand Responsive Transport (DRT) system. This zone encompasses areas where excessive concentration of DRT vehicles can lead to inefficient resource use, and serves to restrict DRT operation. By establishing a Red Zone, DRT services can more intensively deploy resources to vulnerable traffic areas, thereby improving overall operational efficiency.

[0041] Red zone designations are established through a variety of data analysis. Data such as call counts, call success and failure information, call routes, operating routes, station-specific identification information, and spatial information are collected from the DRT operational database (DB). These data are then used to cluster the space and calculate the number of visits. This allows for the identification of DRT demand patterns in specific areas and the designation of high-density demand areas as red zones.

[0042] The primary purpose of establishing red zones is to prevent swamping, where DRT vehicles become trapped in a specific area and are unable to move to other areas. Swamping occurs when a high volume of calls in high-density demand areas prevents DRT vehicles from leaving the area, ultimately preventing them from handling calls in underserved areas. Red zones address this issue and ensure efficient deployment of DRT vehicles.

[0043] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.

[0044] Figure 1 is a reference diagram illustrating an OD (Origin-Destination) pattern of a conventional DRT system to explain the necessity of setting a red zone.

[0045] Referring to Figure 1, we can see the origin-destination (OD) pattern during commuting hours of a conventional DRT system, to which the present invention does not apply. The OD pattern indicates the user's travel route and frequency between their origin and destination, allowing us to identify demand concentrations in specific areas. The data collection period is not limited to commuting hours and can be freely varied according to user settings.

[0046] The OD pattern of the commuting time of a conventional DRT system to which the embodiment of the present invention is not applied is as described below.

[0047] First, we can compare the DRT call success chart (M1) and the DRT call failure chart (M2). The DRT call success chart (M1) shows the top 15 OD patterns with successful DRT calls during commuting hours, while the DRT call failure chart (M2) shows the top 15 OD patterns with failed DRT calls during commuting hours. Comparing the two charts reveals the problem with the existing DRT system: it cannot effectively manage DRT operations between vulnerable areas (B1, B2) and non-vulnerable areas (A1, A2).

[0048] Transportation-disadvantaged areas (A1, A2) have ample public transportation alternatives and high population densities, resulting in a high concentration of DRT calls. As shown in the diagram, the DRT call success chart (M1) shows a concentrated pattern of successful call ODs in the transportation-disadvantaged area (A1). This concentrated pattern demonstrates that significant DRT demand is generated in the transportation-disadvantaged area (A1).

[0049] On the other hand, DRT services play a crucial role in areas with limited public transportation (B1, B2). According to the diagram, the DRT call failure chart (M2) shows a relatively higher number of DRT call failures in the transportation-vulnerable area (B2) during rush hour than in the non-vulnerable area (A2). This reflects the phenomenon of DRT vehicles being concentrated in the transportation-vulnerable area (A2) and failing to accommodate calls from the transportation-vulnerable area (B2).

[0050] As can be seen from the drawing, it can be confirmed that the conventional DRT system is not operating according to its original purpose due to excessive DRT calls in traffic-vulnerable areas (A1, A2), and the necessity of setting a red zone proposed by the present invention to solve this problem can be emphasized.

[0051] FIG. 2 is a schematic diagram schematically illustrating a red zone creation system for operating a DRT platform according to an embodiment of the present invention.

[0052] As illustrated in FIG. 2, the DRT platform (10) according to an embodiment of the present invention may include a DRT operation DB (200), a red zone creation system (100), and a DRT operation system (300).

[0053] The red zone creation system (100) can collect and analyze data required for red zone creation from the DRT operation DB (200) to create a red zone. The DRT operation system (300) can control DRT operation based on the red zone created by the red zone creation system (100). For example, the DRT operation system (300) can prevent resource waste by restricting DRT operation within a red zone and concentrate resources in traffic-vulnerable areas.

[0054] The DRT operation database (200) stores and manages DRT demand data, including the number of DRT calls, call success and failure information, call routes, operating routes, and identification information for each stop. Furthermore, the DRT operation database (200) may include spatial information (geographic information) including real-time location information of DRT vehicles, user waiting times, boarding and alighting times, traffic volume data, and environmental factor data. These various data can be used as basic data for the red zone creation system (100) to efficiently set red zones and optimize DRT services.

[0055] The DRT operating system (300) controls DRT operation based on red zone information received from the red zone generation system (100). By restricting DRT operation within red zones, resource waste is prevented, and by limiting DRT calls within red zones, resources are concentrated in vulnerable traffic areas. This maximizes the operational efficiency of DRT services.

[0056] The user interface (400) provides DRT operation information, including red zone information received from the DRT operation system (300), to the DRT operator or user. Furthermore, the user interface (400) allows the DRT operator to monitor the system in real time and adjust red zone settings.

[0057] The user interface (400) can be implemented in a program or software and can be utilized on various devices. The devices can include mobile devices capable of wired or wireless communication, such as laptops, smartphones, tablet PCs, and PDAs, as well as desktop computers, TVs, monitors, and wearable devices. This allows operators to dynamically change red zones based on real-time data anytime, anywhere, and efficiently manage the DRT system by responding appropriately to situations.

[0058] Additionally, the user interface (400) provides operators with the ability to adjust red zone settings. This allows operators to dynamically adjust red zones based on real-time data and respond appropriately to situations. For example, red zones can be adjusted during specific times or when specific events occur, optimizing DRT operation.

[0059] Figure 3 is a block diagram of a red zone generation system according to an embodiment of the present invention.

[0060] As illustrated in FIG. 3, the red zone generation system (100) may include a DRT demand data collection unit (110), a spatial clustering unit (120), a spatial cluster visit count calculation unit (130), an objective function setting unit (140), a red zone representative stop extraction unit (150), a red zone additional stop extraction unit (160), and a red zone reflection unit (170).

[0061] First, the DRT demand data collection unit (110) can collect various data in real time from the DRT operation DB (200), such as the number of DRT calls, call success and failure information, call path, operation path, identification information for each stop, and spatial information (geographic information).

[0062] Spatial information can include bus stop location information (latitude and longitude), surrounding terrain information, road network information, and route information between bus stops. This allows the DRT system to calculate distances between bus stops, find optimal routes, and analyze call demand. Furthermore, at least some of the collected data is utilized as DRT demand data, playing a crucial role in red zone creation.

[0063] In particular, the DRT demand data collection unit (110) can use demand data collected from the DRT operation database (200) to calculate daily demand data for stations in the analyzed area. The daily demand data collection period is not limited to one day and can be arbitrarily changed. Additionally, the daily demand data can be set as the daily average of demand data over a specific period.

[0064] The spatial clustering unit (120) can cluster spatially similar regions into multiple spatial clusters based on at least one of the collected demand data and spatial information (geographic information). The spatial clustering unit (120) can apply a clustering algorithm to create the spatial clusters. For example, algorithms such as K-means clustering or DBSCAN can be used.

[0065] The process of dividing the analysis area into clusters by analyzing spatial information using DRT demand data will be described later.

[0066] The spatial cluster visit count calculation unit (130) uses spatial cluster information and the aforementioned demand data to calculate the number of visits to each cluster, thereby identifying the DRT demand for that cluster. This allows the spatial cluster visit count calculation unit (130) to identify the visit frequency and pattern of a specific cluster. The detailed process for calculating cluster visit count data based on travel demand will be described later.

[0067] The objective function setting unit (140) can set an objective function for extracting representative red zone stops. For example, the objective function setting unit (140) can set a function that minimizes the variance of the number of visits to a cluster as the objective function. The objective function can be used to select stops that contribute more than a predetermined value to minimizing the variance of the number of visits to spatial clusters.

[0068] The process of setting the objective function will be described in detail later.

[0069] The red zone representative stop extraction unit (150) can extract stops with a contribution greater than a predetermined value to minimize the variance of the objective function Z as red zone representative stops. The red zone representative stops serve as a reference point for setting red zone boundaries and enable optimal resource allocation.

[0070] Modified Bayesian Information Criterion (ModBIC) can be used as an algorithm to extract representative red zone stops. ModBIC is a statistical criterion that balances model adequacy and complexity to select the optimal model. This prevents overfitting, improves generalization, and enables efficient red zone identification. Furthermore, exception conditions can be added to the ModBIC algorithm, which can be flexibly adjusted based on the situation.

[0071] The detailed process of extracting the red zone representative stops that minimize the objective function will be described later.

[0072] The red zone additional stop extraction unit (160) extracts additional stops to be included in the red zone based on the similarity of distance and OD patterns with respect to the representative red zone stop. The red zone additional stops serve to expand the red zone.

[0073] If traffic restrictions are imposed only at the representative red zone stops, traffic may be diverted to nearby stops with similar traffic uses. Therefore, similarity checks are needed to cluster nearby stops with high similarity, centered around the representative stop, into red zones.

[0074] The detailed process of extracting similar additional stops based on the representative red zone stop will be described later.

[0075] The red zone reflection unit (170) can generate a polygonal red zone based on the red zone representative stops and red zone additional stops extracted by the red zone representative stop extraction unit (150) and the red zone additional stop extraction unit (160) and provide the polygonal red zone to the DRT operating system (300). The red zone having the boundary of the polygon generated by the red zone reflection unit (170) can serve as a criterion for restricting DRT operation.

[0076] The polygon generation method of the above red zone boundary can be performed using the Convex Hull algorithm or the Voronoi diagram. The Convex Hull algorithm is a method of generating a convex polygon by connecting the outermost stops, and the Voronoi diagram is a method of defining the influence range of each stop based on the midline between the stops.

[0077] The red zone reflection unit (170) updates the DRT map in real time during DRT system operation, enabling real-time updates of the red zone and their reflection in DRT operation. This allows the DRT system to adapt to dynamically changing traffic conditions.

[0078] In particular, this system is not limited to a single red zone; it can simultaneously create and manage multiple red zones, enabling efficient resource allocation and operational control in both vulnerable and vulnerable areas. This allows the DRT system to provide customized transportation services that reflect diverse regional characteristics.

[0079] Operators can monitor the boundaries of the red zone through the user interface (400) illustrated in Figure 2 and adjust them as needed. This allows for flexible responses to changing traffic conditions. Furthermore, DRT vehicles operate based on updated maps, and operation within red zones is restricted. The DRT map is updated periodically or in real time, ensuring it is always up-to-date.

[0080] Calls between P32 and P33 are possible. Similarly, DRT calls between stations within a cluster, such as P12 and P32, are possible. However, DRT calls between stations within the red zone, such as P11 and P21, are restricted. In other words, by restricting calls within the red zone and allowing calls to stations outside the red zone, the swamp phenomenon occurring within the red zone can be prevented.

[0081] FIG. 5 is a flowchart illustrating a red zone creation process for operating a DRT platform according to an embodiment of the present invention.

[0082] In step (S110), the red zone creation system for DRT platform operation collects various data in real time from the DRT operation database, including the number of DRT calls, call success and failure information, call routes, operation routes, identification information for each stop, and spatial information. This data is used as basic data for red zone creation.

[0083] In step (S120), the red zone creation system for DRT platform operation analyzes spatial information based on daily demand data from stops in the analysis area, divides the analysis area into k spatial clusters, and defines attribute vectors P, S, and R for individual stops and a matrix D representing the number of calls generated between stops. Spatial clusters can be defined by considering the spatial distribution and proximity of stops.

[0084] Below, we mathematically describe the process of dividing the analysis area into clusters by analyzing spatial information using DRT demand data.

[0085] First, P: Bus stops existing within the analysis area

[0086] Vector P: A set of all n stops in the analysis area

[0087]

[0088] The spatial cluster to which a stop p belongs can be determined as k based on the accessibility between the stops. Accessibility can be determined using the spatial distribution and proximity between the stops.

[0089] Vector S: A set of numbers of the spatial cluster to which station p belongs

[0090]

[0091] Vector R: A set of masked elements that are 1 if stop p is in the red zone, and 0 otherwise.

[0092]

[0093] Matrix D: A matrix defining the number of calls occurring between stops during a specific time unit (1 day) (rows: boarding stops, columns: alighting stops)

[0094]

[0095] d ij : Stop p for a specific time unit (1 day) i Get on at stop p j Total number of calls that were dropped off

[0096] d ij = demand from i to j

[0097] The mathematical definitions disclosed above define relevant data to create red zones, including bus stops, clusters, and call information within the analysis area, and serve as the basis for a methodology to evaluate and optimize travel demand within and outside the red zone based on this data.

[0098] In step (S130), the red zone creation system for DRT platform operation calculates the number of visits to spatial clusters. This allows the visit frequency and pattern of each cluster to be identified, and the spatial cluster visit count calculation unit then evaluates the DRT demand for each cluster.

[0099] In step (S140), the red zone generation system for operating the DRT platform sets the variance of cluster visit counts as an objective function. Red zone stops can be selected by minimizing this objective function, i.e., the variance of the visit counts of spatial clusters. Specifically, the variance, which is the objective function, is calculated by squaring the differences between the number of visits for each cluster and the overall average number of visits and then summing them. By minimizing this variance, DRT resources can be prevented from being overly concentrated or under-concentrated in specific clusters, and overall resource allocation can be balanced. However, since unconditionally minimizing the objective function does not allow for the selection of meaningful red zone stops, red zone stops that contribute to minimizing the objective function can be selected. A method for selecting red zone stops based on the objective function will be described later with reference to FIG. 6.

[0100] In step (S150), the red zone generation system for DRT platform operation extracts a representative red zone stop that minimizes the objective function. This representative red zone stop serves as the center of the red zone and serves as the standard for DRT operation. The specific method for selecting this representative red zone stop is described later in Figure 6.

[0101] In step (S160), the red zone creation system for DRT platform operation can extract additional similar stops based on the representative red zone stop. These additional stops can serve to expand the red zone and restrict DRT operation within the red zone. The method for selecting similar red zone stops is described in detail below with reference to FIGS. 7 through 9.

[0102] In step (S170), the red zone creation system for DRT platform operation can reflect red zones on the DRT map based on the representative and additional stops from which the red zone reflection area is extracted. This can restrict DRT operation within the red zone.

[0103] FIG. 6 is a flowchart illustrating a process of setting an objective function and extracting a red zone representative stop in a red zone creation method for operating a DRT platform according to an embodiment of the present invention.

[0104] In step (S141), the red zone generation system can calculate a matrix of calls generated between stops, excluding calls between red zone stops. Below, the process of constructing the matrix based on call data between the remaining stops, excluding call data between stops within the red zone, is mathematically explained.

[0105] D ij : Original call demand from stop i to stop j

[0106] R i ,R j : An indicator indicating whether stops i and j are in the red zone. 1 if they are in the red zone, 0 if they are not.

[0107] : Matrix D, excluding the passage between stops included in the red zone

[0108] : Call matrix value considering the red zone from stop i to stop j in the mth iteration

[0109] [Mathematical Formula 1]

[0110]

[0111] In mathematical expression 1 becomes 1 when both stops are in the red zone, otherwise it becomes 0. Therefore, If both stops are not in the red zone or only one of them is in the red zone, it is 1, and if both stops are in the red zone, it is 0. That is, if both stops are in the red zone, the call demand is 0, otherwise, the original call demand D ij is maintained.

[0112] In step (S142), the red zone generation system can calculate a matrix of the number of cluster visits due to traffic at each stop. This matrix can indicate how frequently each stop is visited based on the cluster to which it belongs.

[0113]

[0114] First, SD k : The number of visits to spatial cluster k according to traffic demand at a specific point in time

[0115] : Average number of visits at that time

[0116] Q k : A matrix consisting of Kronecker delta functions representing the stops belonging to each cluster

[0117]

[0118] : Kronecker delta function, defined as 1 when the station belongs to cluster k, otherwise 0.

[0119]

[0120] : Matrix Q contains information about the cluster to which each stop belongs, is the transposed matrix.

[0121] SDI: SDI quantitatively represents the demand for a specific stop by assessing how frequently each stop is visited. It is an n×k matrix (rows: stop index, columns: spatial cluster index). For example, the value in row i and column k represents the number of visits to cluster k due to trips to stop i.

[0122] The number of visits to spatial cluster k according to traffic demand at a given point in time is SD k The function for calculating is as shown in mathematical formula 2.

[0123] [Equation 2]

[0124]

[0125] SD k To calculate the SDI of the m-th iteration (m) SDI (Spatial Demand Index) is an index that quantitatively evaluates the demand for visiting each stop for efficient operation of the DRT system. SDI (m) may include the diag(M) function, which is a function that extracts the diagonal elements of a matrix. SDI (m) is determined by mathematical formula 3.

[0126] [Equation 3]

[0127]

[0128] represents the SDI value for a specific stop i in the m-th iteration.

[0129] For example, suppose there are stops A, B, and C, and the SDI values ​​of each stop are 10, 15, and 20, respectively.

[0130] That is, SDI[A]=10, SDI[B]=15, SDI[C]=20.

[0131] Assume that stops A and B belong to cluster k.

[0132] Number of visits to cluster k is calculated as follows:

[0133]

[0134] The number of visits to cluster k is 25.

[0135] In step (S143), the red zone generation system can repeatedly perform calculations to minimize an objective function. In this step, calculations can be performed to find a representative red zone stop that minimizes the variance of the number of cluster visits.

[0136] The objective function is defined as follows:

[0137] [Equation 4]

[0138]

[0139] : Objective function value at m-th iteration

[0140] : Number of visits to spatial cluster k in the m-th iteration

[0141] : Average number of visits to all clusters in the m-th iteration

[0142] Below, we present a mathematical solution to the process of calculating cluster visit count data according to traffic demand and the process of minimizing the objective function Z based on the spatial cluster visit data according to demand.

[0143] The formula for minimizing the objective function is as shown in Equation 5.

[0144] [Equation 5]

[0145]

[0146] Mathematical expression 5 calculates the variance by adding the squares of the differences between the number of visits to each cluster and the average number of visits.

[0147] In step (S144), the red zone generation system searches for the stops that have the greatest impact on minimizing the objective function, i.e., the highest contribution, and ultimately extracts a representative red zone stop. This stop serves as the center of the red zone and can serve as a benchmark for DRT operation.

[0148] The red zone generation system uses the following mathematical expression 6 to minimize the objective function Z by selecting the most influential stop I. p can be decided.

[0149] The most influential stop I to minimize the objective function Z p The formula to determine is as follows:

[0150] [Equation 6]

[0151]

[0152] : Number of visits to clusters

[0153] : Number of visits to clusters s including stop p

[0154] Mathematical expression 6 represents the difference in the objective function value when a specific stop p is included in the red zone and when it is not included. That is, I p This means calculating the change in the objective function Z when the stop p is included in the red zone and selecting the stop that makes this change large.

[0155] At step (S145), the red zone generation system can check the set conditions to determine whether additional calculations are necessary. If the conditions are met, the calculation can proceed again, including the added red zone stop. If the conditions are not met, step (S150) is performed.

[0156] Additionally, exception conditions can be introduced during the decision-making process. Exception condition (1) allows the DRT system to reliably select red zone stops during the initial learning phase. Furthermore, exception condition (2) allows the DRT system to maintain system efficiency by preventing excessive demand reduction.

[0157] Below, the mathematical solution to the setting conditions and exception conditions of the red zone stop is disclosed.

[0158] [Equation 7]

[0159]

[0160] n: number of stops

[0161] : The stop I that has the greatest influence in the m-th iteration p The influence of

[0162] nrr: degree of freedom of the model

[0163] Based on mathematical formula 7, In this case, select the corresponding stop as the red zone.

[0164] In addition, if the region p is selected as a red zone according to the exceptional conditions or if the exceptional conditions are met, DR (m) Repeat the process again from the calculation step.

[0165] ModBIC's exception conditions are as follows:

[0166] Exception condition (1): When the number of iterations is less than 2

[0167] Exception condition (2): If the total demand reduction due to the entire Red Zone is less than α% of the initial total demand (0 < α < 100).

[0168] In step (S146), the red zone generation system can recalculate the number of calls between stops, including the added red zone stop. By additionally defining stop p as a red zone Update.

[0169] In step (S150), the red zone generation system is based on the results calculated in the previous iteration. The stop index with a value of 1 is defined as the representative stop in the red zone. This stop serves as the center of the red zone and serves as an important criterion for red zone expansion and additional stop selection in subsequent iterations. This ensures consistency in red zone settings.

[0170] represents the red zone stop vector at iteration m-1. This vector indicates whether each stop is included in the red zone, and has a value of 1 if it is included in the red zone and 0 if it is not. Therefore, We can find the stop whose definition value is 1 and define that stop as the representative stop of the red zone for this iteration based on its index.

[0171] FIG. 7 is a flowchart illustrating a process of extracting additional red zone stops and determining a red zone area in a red zone creation method for operating a DRT platform according to an embodiment of the present invention.

[0172] In step (S161), the red zone generation system can calculate the similarity between the representative red zone stop and a regular stop. This similarity calculation can be performed using a cost function and an OD pattern similarity function. The cost function calculates the travel distance between the representative stop and a regular stop, and the OD pattern similarity function calculates the similarity of the OD patterns between the representative stop and a regular stop.

[0173] Below, the process of evaluating the similarity between the selected red zone representative stop r and the remaining general stops i using the cost function and the OD pattern similarity function is mathematically explained and disclosed.

[0174] β: A criterion constant that determines similarity, 0<β<1

[0175] γ: Threshold constant for similarity evaluation, 0<γ<1

[0176] x: distance between representative red zone stop r and general stop i

[0177] d rs : The number of calls from a representative red zone stop r to a specific OD element s (a specific destination)

[0178] d r : Total number of calls for the entire OD pattern at the representative red zone stop.

[0179] d is : Number of calls from a general stop i to a specific OD element s

[0180] d i : Total number of calls for the entire OD pattern at a general stop

[0181] The function that evaluates the cost for the travel distance between the representative red zone stop r and the general stop i is as shown in mathematical expression 8.

[0182] [Equation 8]

[0183]

[0184] At this time, the exponent part of mathematical expression 8, -0.0095(x-500), is only an example, and the sigmoid function can be customized according to the characteristics of the red zone application area or the operator's settings.

[0185]

[0186] The function for evaluating the OD pattern between the representative red zone stop r and the general stop i is as shown in mathematical expression 9.

[0187] [Equation 9]

[0188]

[0189] The similarity evaluation between the representative red zone stop r and the general stop i is determined by mathematical expression 10.

[0190] [Equation 10]

[0191]

[0192] In step (S162), the red zone generation system can search for general stops that satisfy a similarity threshold set based on the similarity evaluation results. Similarity is evaluated based on a value obtained by combining a cost function and an OD pattern similarity function using a weight β. If the similarity evaluation value exceeds a specific threshold γ, the stop is selected as an additional red zone stop.

[0193] The similarity between the selected red zone representative stop and the remaining general stops is evaluated, and if the similarity value is 1, the stop can be included in the red zone.

[0194] In step (S163), the red zone creation system may designate the discovered stops based on similarity as additional red zone stops. The designated additional red zone stops are used to set the boundaries of the red zone in the DRT system.

[0195] In step (S170), the red zone generation system can determine the red zone area by generating a red zone polygonal surface for all red zone stations, including the red zone representative station and additional stations. The red zone area can be generated using a convex hull algorithm or a Voronoi diagram. The generated red zone polygonal surface can include or replace surfaces, lines, or points, and serves as a criterion for restricting DRT operation within the red zone.

[0196] FIG. 8 is a reference diagram illustrating a method for evaluating distance similarity in a red zone creation method for operating a DRT platform according to an embodiment of the present invention.

[0197] According to Figure 8, the first section schematically illustrates the locational relationship between the red zone representative stop and the regular stop. Here, the black circle represents the red zone representative stop (r), and the white diamond represents the regular stop (i). The red zone is indicated by a red dotted line and is set at a 600-meter radius. In this diagram, the distance between the red zone representative stop (r) and the regular stop (i) is indicated as 600 meters.

[0198] The second part is a graph representing the cost function for evaluating distance similarity. The horizontal axis represents the distance (in meters) between the representative red zone stop r and the regular stop i, and the vertical axis represents the value of the cost function Cost(r, i). As can be seen in the graph, the cost function value increases as the distance between the two stops increases. The function that evaluates the cost of the distance between the representative red zone stop r and the regular stop i can be expressed by Equation 8 above.

[0199] For example, the value of the cost function at a specific distance (600 m) is as follows.

[0200]

[0201] FIG. 9 is a reference diagram illustrating a method for evaluating OD pattern similarity in a red zone creation method for operating a DRT platform according to an embodiment of the present invention.

[0202] This method analyzes demand generated at red zone representative stops (r) and regular stops (i) by classifying them into spatial clusters. The diagram is largely divided into two parts.

[0203] The first part shows the call occurrence rate from the red zone representative stop r to the spatial cluster.

[0204] Spatial cluster 1: 25%

[0205] Spatial cluster 2: 60%

[0206] Spatial cluster 3: 15%

[0207] The second part shows the call occurrence rate from the red zone general stop i to the spatial cluster.

[0208] Spatial cluster 1: 15%

[0209] Spatial cluster 2: 70%

[0210] Spatial cluster 3: 15%

[0211]

[0212] To assess the OD pattern similarity between two stations, the difference in occurrence rates for each spatial cluster is calculated. The function for assessing OD pattern similarity can be expressed as Equation 9 above. The solution process using Equation 9 is as follows.

[0213]

[0214] Below, the process of determining similarity based on the values ​​obtained in FIGS. 8 and 9 is mathematically explained and disclosed.

[0215] First, the formula for similarity check is as shown in Equation 10 above.

[0216] for example, When defined as , the conditions for the similarity value to be 1 are as follows.

[0217]

[0218] If the values ​​calculated in Figures 8 and 9 are substituted into the above conditions, the results are as follows.

[0219] (1) 0.7×(1-0.72)=0.196

[0220] (2) 0.3×(0.876)=0.2628

[0221] 0.196+0.2628=0.4588

[0222] ∴0.4588<0.65

[0223] Therefore, since the given condition is not satisfied, the value of the Similarity function becomes 0, leading to the conclusion that there is no similarity.

[0224] Similarity(r,i)=0

[0225] The red zone creation system of the present invention can efficiently manage DRT system resources and improve service quality in traffic-vulnerable areas. This reduces resource waste and maximizes the overall operational efficiency of the DRT system.

[0226] The foregoing description of the present invention is for illustrative purposes only, and those skilled in the art will readily appreciate that the present invention can be readily modified into other specific forms without altering the technical spirit or essential characteristics of the present invention. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, each component described as a single entity may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined manner.

[0227] The scope of the present invention is indicated by the claims set forth below, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention.

[0228] The mode for carrying out the invention is described together with the best mode for carrying out the invention.

[0229] According to an embodiment of the present invention, red zones can be created to limit the operating area of ​​DRT vehicles. This can guarantee the right to mobility for citizens in vulnerable transportation areas by limiting calls between red zone stops.

[0230] According to an embodiment of the present invention, variable settings can be changed based on environmental conditions and the operator's judgment. This allows for efficient operation of the DRT system based on environmental conditions by variably selecting the red zone.

Claims

1. In the red zone creation system for operating the DRT platform, A DRT demand data collection unit that collects DRT demand data in real time, including at least one of the number of DRT calls, call success and failure information, call route, operation route, stop-by-stop identification information, and spatial information; A spatial clustering unit that forms spatial clusters by clustering stops in the analysis area based on the above DRT demand data and spatial information; A spatial cluster visit count calculation unit that calculates the number of times each stop visits each cluster; An objective function setting unit that sets an objective function based on the number of visits to the above spatial cluster; A red zone representative stop extraction unit that extracts a red zone representative stop based on the degree to which it contributes to minimizing the above objective function; A red zone additional stop extraction unit that selects additional stops based on the similarity judgment result based on the extracted red zone representative stops; and A red zone reflection section that sets a red zone area based on the extracted red zone representative stops and red zone additional stops and reflects it on the DRT map; Including, A red zone generation system that minimizes the above objective function by minimizing the variance of the number of visits of each cluster to each of the above stations.

2. In paragraph 1, A red zone generation system in which the above objective function setting unit sets an objective function that calculates a variance value by adding the square of the difference between the number of visits to each cluster and the average number of visits.

3. In paragraph 2, The red zone representative stop extraction unit calculates a change in an objective function when the first stop is included in the red zone, selects a stop that minimizes this change, and, if additional calculation is required based on predetermined conditions, includes the first stop as an additional red zone stop to perform additional calculation. A red zone generation system.

4. In paragraph 3, A red zone generation system in which the above predetermined conditions are determined using ModBIC (Modified Bayesian Information Criterion).

5. In paragraph 4, A red zone generation system including a red zone additional stop extraction unit that determines similarity between stops based on an evaluation of OD (Origin-Destination) pattern similarity or cost for distance.

6. In paragraph 5, A red zone generation system including setting the DRT platform to restrict DRT calls within the set red zone and allow DRT calls to stops outside the red zone.

7. In the method of creating a red zone for the DRT platform operation, the red zone creation system: A step of collecting DRT demand data in real time, including at least one of the number of DRT calls, call success and failure information, call path, operation path, stop-by-stop identification information, and spatial information; A step of forming a spatial cluster by clustering stops in the analysis area based on the above DRT demand data and spatial information; A step of calculating the number of times each stop visits each cluster; A step of setting an objective function based on the number of visits to the above spatial clusters; A step of extracting a red zone representative stop based on the degree to which it contributes to minimizing the above objective function; A step of selecting additional stops based on the similarity judgment result based on the extracted red zone representative stops; and A step of setting a red zone area based on the extracted red zone representative stop and the red zone additional stop and reflecting it on a DRT map; Including, A red zone creation method in which minimizing the above objective function is to minimize the variance of the number of visits to each cluster by each stop.

8. In paragraph 7, A method for creating a red zone, wherein the step of setting the above objective function is to set an objective function that calculates a variance value by adding the squares of the differences between the number of visits to each cluster and the average number of visits.

9. In paragraph 8, The step of extracting the above red zone representative stop is a method for creating a red zone, wherein the step of calculating the change in the objective function when the first stop is included in the red zone, selecting a stop that greatly increases this change, and, if additional calculation is required based on predetermined conditions, including the first stop as an additional red zone stop, performing additional calculation.

10. In paragraph 9, A red zone generation system in which the above predetermined conditions are determined using ModBIC (Modified Bayesian Information Criterion).

11. In paragraph 10, A method for generating a red zone, wherein the step of extracting the above red zone additional stops includes judging the similarity between stops based on an evaluation of the OD (Origin-Destination) Pattern similarity or cost for distance.

12. In paragraph 11, A method for creating a red zone, wherein the step of setting the above red zone area and reflecting it on the DRT map includes setting the DRT platform to restrict DRT calls within the set red zone and allow DRT calls to stops outside the red zone.

Images