Transportation plan management device and transportation plan management method
The transportation plan management device analyzes causal relationships to improve KPIs by identifying inconsistent KPIs and generating revised plans, addressing the complexity of transportation service management and improving plan efficiency.
Patent Information
- Application Number
- PCT/JP2024/001401
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-22
- Filing Date
- 2024-01-19
- Publication Date
- 2024-11-28
AI Technical Summary
Conventional transportation plan management systems struggle to identify and address the underlying reasons for Key Performance Indicators (KPIs) not meeting required thresholds, particularly in complex transportation services, due to the lack of clear causal relationships between variables and KPIs, making it difficult to revise plans effectively.
A transportation plan management device and method that includes a processor, memory, and storage, which analyzes existing transportation plans and KPI information to identify inconsistent KPIs, determines causal relationships with related variables, and generates revised plans to meet threshold requirements by extracting relevant statistical information using dimensional parameters.
Enables the creation of revised transportation plans that improve KPIs by understanding the causal relationships between variables and KPIs, facilitating more efficient and effective transportation service management.
Smart Images

Figure JP2024001401_28112024_PF_FP_ABST
Abstract
Description
Transportation plan management device and transportation plan management method
[0001] The present invention relates to a transportation plan management device and a transportation plan management method.
[0002] In transportation services, transportation planning is used to manage the departure and arrival times of trains, buses, etc. in a transportation network, taking into account operational costs and demand. Traditionally, transportation planning has been done manually by transportation operators, taking into account the transportation network infrastructure and passenger needs.
[0003] In recent years, in order to realize a sustainable society, transportation companies such as railway companies and bus companies are required to formulate flexible transportation plans for efficient transportation while taking into consideration fluctuations in demand for transportation services and the impact on the environment. Transportation plans themselves have traditionally been used to manage the movement of moving objects such as trains and buses. For example, Japanese Patent Laid-Open Publication No. 2019-092347 (Patent Document 1) discloses an automatic driving control device that controls trains according to a transportation plan.
[0004] Patent Document 1 describes the technology as follows: "The automatic driving control device includes a creation unit 104, a calculation unit 101, and a control unit 105. The creation unit creates a running plan including a target speed of the train in a specified section. The calculation unit calculates the train acceleration that will make the train's running speed in the specified section the target speed based on at least one of the air resistance and route conditions of the specified section. The control unit controls the acceleration and deceleration of the train in the specified section in accordance with the calculated train acceleration."
[0005] Japanese Patent Application Laid-Open No. 2019-092347
[0006] The automatic driving control device described in Patent Document 1 controls the acceleration and deceleration of a train based on air resistance and route conditions so that the train's running speed reaches a target speed defined in a running plan. The automatic driving control device described in Patent Document 1 is said to be able to improve train ride comfort and energy saving effects.
[0007] Meanwhile, a method for evaluating the performance of transportation plans used in transportation services is sometimes based on key performance indicators (KPIs). These KPIs are quantitative indicators that indicate the performance of transportation services with respect to various factors, such as the operating costs, expected revenue, environmental impact, and energy efficiency of a given transportation plan. If a particular KPI in a transportation plan does not meet a predetermined required threshold, it is desirable to revise the transportation plan so that the KPI meets the required threshold.
[0008] However, while conventional KPI-based evaluation methods can identify KPIs that do not meet the required threshold, the reason why the KPI does not meet the required threshold is unclear, making it difficult to determine how to revise the transportation plan so that the KPI meets the required threshold. In particular, in complex situations such as transportation services, where many variables are involved, such as the number of transportation routes, congestion level, and passenger waiting time, understanding the causal relationships between these variables and KPIs and appropriately revising the existing transportation plan requires a great deal of specialized knowledge, which places a heavy burden on users who manage the transportation plan.
[0009] According to the means described in Patent Document 1, trains can be controlled using a transportation plan that takes into account specific performance indicators such as energy efficiency, but it does not describe a means for proposing specific revisions to improve KPIs while taking into account the causal relationships between numerous variables and KPIs.
[0010] Therefore, the present disclosure aims to provide a transportation plan management means that can generate a revised transportation plan with improved KPIs by taking into account the causal relationships between numerous variables and KPIs in transportation plans for transportation services and analyzing the reasons why KPIs do not meet required thresholds.
[0011] In order to solve the above-mentioned problems, one representative transportation plan management device of the present invention includes a processor, a memory, and a storage unit. The storage unit includes existing transportation plan information that indicates an existing transportation plan for a specified transportation service, transportation service KPI information that includes KPIs that indicate performance indicators for the existing transportation plan information, and statistical information related to the transportation service. The memory includes processing instructions for causing the processor to function as an inconsistent KPI identification unit that identifies inconsistent KPIs among the KPIs that do not meet a specified required threshold based on the existing transportation plan information and the transportation service KPI information, a causal relationship determination unit that analyzes the inconsistent KPIs and the statistical information and determines a causal relationship between the inconsistent KPIs and related variables in a subset of statistical information extracted from the statistical information, and a transportation plan generation unit that generates revised transportation plan information in which the inconsistent KPIs meet the required threshold using the existing transportation plan information and the determined causal relationship.
[0012] According to the present disclosure, it is possible to provide a transportation plan management means that can analyze the reasons why a KPI does not meet a required threshold while taking into account the causal relationships between numerous variables and KPIs with respect to transportation plans for transportation services, and then generate a revised transportation plan that improves the KPI. Other problems, configurations, and effects will be made clear in the description of the following embodiments of the invention.
[0013] FIG. 1 is a diagram illustrating a computer system for implementing an embodiment of the present disclosure. FIG. 2 is a diagram illustrating an example of a configuration of a transportation plan management system according to an embodiment of the present disclosure. FIG. 3 is a diagram illustrating the logical configuration of functional units and databases included in a transportation plan management device according to an embodiment of the present disclosure. FIG. 4 illustrates an overall flow of a transportation plan management method according to an embodiment of the present disclosure. FIG. 5 is a diagram illustrating an example of a flow of a causal relationship determination process according to an embodiment of the present disclosure. FIG. 6 is a diagram illustrating an example of a flow of a causal relationship analysis process according to an embodiment of the present disclosure. FIG. 7 is a diagram illustrating an example of a flow of a transportation plan generation process according to an embodiment of the present disclosure. FIG. 8 is a diagram illustrating an example of a case where a time parameter L according to an embodiment of the present disclosure is set as a dimension parameter. FIG. 9 is a diagram illustrating an example of a case where a vehicle parameter M according to an embodiment of the present disclosure is set as a dimension parameter. FIG. 10 is a diagram illustrating an example of a case where a station parameter N according to an embodiment of the present disclosure is set as a dimension parameter. FIG. 11 is a diagram illustrating an example of a causal relationship diagram according to an embodiment of the present disclosure. FIG. 12 is a diagram illustrating a dimension parameter setting interface according to an embodiment of the present disclosure.
[0014] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. However, the present invention is not limited to this embodiment. In the description of the drawings, the same parts are designated by the same reference numerals. Furthermore, although terms such as "first," "second," and "third" may be used to describe various elements or components in this disclosure, it will be understood that these elements or components should not be limited by these terms. These terms are used only to distinguish one element or component from another. Therefore, a first element or component discussed below could also be referred to as a second element or component without departing from the teachings of the inventive concept.
[0015] 1, a computer system 100 for implementing embodiments of the present disclosure will be described. The mechanisms and devices of various embodiments disclosed herein may be applied to any suitable computing system. Major components of the computer system 100 include one or more processors 102, memory 104, a terminal interface 112, a storage interface 113, an I / O (input / output) device interface 114, and a network interface 115. These components may be interconnected via a memory bus 106, an I / O bus 108, a bus interface unit 109, and an I / O bus interface unit 110.
[0016] Computer system 100 may include one or more general-purpose programmable central processing units (CPUs) 102A and 102B, collectively referred to as processors 102. In some embodiments, computer system 100 may include multiple processors, while in other embodiments, computer system 100 may be a single CPU system. Each processor 102 executes instructions stored in memory 104 and may include an on-board cache.
[0017] In some embodiments, memory 104 may include random-access semiconductor memory, storage devices, or storage media (either volatile or non-volatile) for storing data and programs. Memory 104 may store all or part of the programs, modules, and data structures that implement the functions described herein. For example, memory 104 may store transportation plan management application 150. In some embodiments, transportation plan management application 150 may include instructions or descriptions that execute the functions described below on processor 102.
[0018] In some embodiments, transportation planning management application 150 may be implemented in hardware via semiconductor devices, chips, logic gates, circuits, circuit cards, and / or other physical hardware devices instead of or in addition to a processor-based system. In some embodiments, transportation planning management application 150 may include data other than instructions or descriptions. In some embodiments, cameras, sensors, or other data input devices (not shown) may be provided to communicate directly with bus interface unit 109, processor 102, or other hardware in computer system 100.
[0019] Computer system 100 may include a bus interface unit 109 that facilitates communication between processor 102, memory 104, display system 124, and I / O bus interface unit 110. I / O bus interface unit 110 may couple to an I / O bus 108 for transferring data to and from various I / O units. I / O bus interface unit 110 may communicate via I / O bus 108 with multiple I / O interface units 112, 113, 114, and 115, also known as I / O processors (IOPs) or I / O adapters (IOAs).
[0020] Display system 124 may include a display controller, a display memory, or both. The display controller may provide video, audio, or both data to display device 126. Computer system 100 may also include one or more sensors or other devices configured to collect data and provide the data to processor 102.
[0021] For example, computer system 100 may include biometric sensors that collect heart rate data, stress level data, etc., environmental sensors that collect humidity data, temperature data, pressure data, etc., and motion sensors that collect acceleration data, movement data, etc. Other types of sensors may also be used. Display system 124 may be connected to a display device 126, such as a standalone display screen, a television, a tablet, or a handheld device.
[0022] The I / O interface unit provides functionality for communicating with various storage or I / O devices. For example, the terminal interface unit 112 may be attached to user I / O devices 116, such as user output devices such as a video display, a television with speakers, and user input devices such as a keyboard, a mouse, a keypad, a touchpad, a trackball, buttons, a light pen, or other pointing device. A user may use a user interface to enter input data or instructions into the user I / O devices 116 and the computer system 100, and receive output data from the computer system 100, by operating the user input devices. The user interface may be displayed on a display, played through speakers, or printed via a printer via the user I / O devices 116, for example.
[0023] Storage interface 113 may accept one or more disk drives or direct access storage devices 117 (typically magnetic disk drive storage devices, but may also be an array of disk drives or other storage devices configured to appear as a single disk drive). In some embodiments, storage device 117 may be implemented as any secondary storage device. The contents of memory 104 may be stored in storage device 117 and retrieved as needed from storage device 117. I / O device interface 114 may provide an interface to other I / O devices, such as printers, fax machines, etc. Network interface 115 may provide a communications path that allows computer system 100 and other devices to communicate with each other. This communications path may be, for example, network 130.
[0024] In some embodiments, computer system 100 may be a device that receives requests from other computer systems (clients) without a direct user interface, such as a multi-user mainframe computer system, a single-user system, or a server computer. In other embodiments, computer system 100 may be a desktop computer, a portable computer, a laptop, a tablet computer, a pocket computer, a telephone, a smartphone, or any other suitable electronic device.
[0025] Next, a transportation plan management system according to an embodiment of the present disclosure will be described with reference to FIG.
[0026] FIG. 2 is a diagram illustrating an example of the configuration of a transportation plan management system 200 according to an embodiment of the present disclosure. The transportation plan management system 200 is a system for generating a revised transportation plan in which the KPI is improved after analyzing the reason why the KPI does not meet a required threshold while taking into account the causal relationships between numerous variables and KPIs regarding a transportation plan for a transportation service. As shown in FIG. 2 , the transportation plan management system 200 includes a transportation plan management device 210, a communication network 250, and a user terminal 260. In the transportation plan management system 200, the transportation plan management device 210 and the user terminal 260 may be connected to each other via the communication network 250.
[0027] The transportation plan management device 210 is a device for generating a revised transportation plan in which inconsistent KPIs that do not satisfy a predetermined required threshold in an existing transportation plan are revised so that the inconsistent KPIs satisfy the required threshold, and as shown in Fig. 2, mainly includes a memory 220, a storage unit 230, a processor 244, and an input / output unit 246. In an embodiment, the transportation plan management device 210 may be implemented by the computer system 100 shown in Fig. 1.
[0028] The memory 220 may be a memory for storing a transportation plan management application 150 for implementing the functions of the transportation plan management means according to an embodiment of the present disclosure. The transportation plan management application 150 may include processing instructions for implementing the functions of software modules such as a KPI analysis unit 222, a causal relationship management unit 224, a revision proposal unit 226, and a transportation plan management unit 228, as shown in FIG.
[0029] The KPI analysis unit 222 is a functional unit that identifies inconsistent KPIs that do not satisfy a predetermined required threshold based on existing transportation plan information and transportation service KPI information stored in the transportation service information DB 232 (described later), and extracts information (a subset of statistical information) for determining causal relationships related to the identified inconsistent KPIs from the transportation service information DB 232 and the local community information DB 234. Note that the details of the operation of the KPI analysis unit 222 will be described later, and therefore will not be described here.
[0030] The causal relationship management unit 224 is a functional unit that determines the causal relationship that exists between the inconsistent KPI identified by the KPI analysis unit 222 and the related variables in the extracted subset of statistical information, and generates inconsistency hypothesis information that indicates the reason why the inconsistent KPI does not satisfy the required threshold. Note that the details of the operation of the causal relationship management unit 224 will be described later, so the explanation thereof will be omitted here.
[0031] The revision proposal unit 226 is a functional unit that generates variable change operation information indicating a recommended variable change operation for a related variable that is in a causal relationship with the inconsistency KPI, based on the inconsistency hypothesis information determined by the causal relationship management unit 224, and calculates the amount of change in the inconsistency KPI when the variable change operation is performed. Note that the details of the operation by the revision proposal unit 226 will be described later, so their explanation will be omitted here.
[0032] The transportation plan management unit 228 is a functional unit for generating revised transportation plan information that uses existing transportation plan information and the determined causal relationships to correct the inconsistent KPIs so that they satisfy the required thresholds. Note that the details of the operation of the transportation plan management unit 228 will be described later, and therefore will not be described here.
[0033] The memory unit 230 is a memory area that contains a database (hereinafter referred to as "DB") for storing various information related to an embodiment of the present disclosure, and may include a transportation service information DB 232 and a local community information DB 234, as shown in FIG. 2.
[0034] The transportation service information DB 232 is a database for storing transportation service information related to transportation services. The transportation service information stored in the transportation service information DB 232 may be, for example, information obtained from a transportation company that operates the transportation service. The local community information DB 234 is a database for storing local community information related to the local community in which the transportation service is provided. The local community information DB 234 may be obtained from any third party, such as a local government, an administrative agency, a company, or an organization, or may be obtained by sensors installed in the community. Note that FIG. 2 shows an example of a configuration in which the transportation service information DB 232 and the local community information DB 234 are stored in the storage unit 230 of the transportation plan management device 210. However, in reality, the information stored in the transportation service information DB 232 and the local community information DB 234 may be stored in an external server (e.g., a server owned by a transportation company or a server of a local government) communicatively connected to the transportation plan management device 210 and may be obtained as needed via the communication network 250.
[0035] The processor 244 is a processing unit for executing processing instructions stored by the memory 220 that define the functionality of each functional unit of the transportation planning management application 150 .
[0036] The input / output unit 246 is a functional unit for accepting information input to the transportation plan management device 210 and outputting data such as revised transportation plan information generated by the transportation plan management device 210. In an embodiment, the input / output unit 246 may include, for example, a keyboard, a mouse, a display that displays a GUI (Graphical User Interface), etc. In an embodiment, the input / output unit 246 may provide the user terminal 260 with a GUI that inputs and outputs various types of information.
[0037] Communications network 250 may include, for example, a local area network (LAN), a wide area network (WAN), a satellite network, a cable network, a WiFi network, or any combination thereof.
[0038] The user terminal 260 is a terminal device that can be used by a user of the transportation plan management device 210. By using the user terminal 260, the user can input information such as existing transportation plan information, transportation service KPI information, and dimension parameters described below, and can check information such as revised transportation plan information output by the transportation plan management device 210. As an example, the user terminal 260 may include, for example, a smartphone, smart watch, tablet, or PC of a user who manages transportation plans at a transportation company, and is not particularly limited. Note that, for convenience of explanation, FIG. 2 illustrates an example of a configuration including one user terminal 260, but the number of user terminals 260 is not limited, and a configuration including multiple user terminals 260 is also possible.
[0039] According to the transportation plan management system 200 described above, it is possible to analyze the reasons why a KPI does not meet the required threshold while taking into account the causal relationships between numerous variables and KPIs for transportation plans for transportation services, and then generate a revised transportation plan that improves the KPI.
[0040] Next, with reference to FIG. 3, a logical configuration of functional units and databases included in the transportation plan management device according to the embodiment of the present disclosure will be described.
[0041] FIG. 3 is a diagram showing the logical configuration of functional units and databases included in the transportation plan management device 210 according to an embodiment of the present disclosure.
[0042] As described above, the transportation plan management device 210 includes two databases, the transportation service information DB 232 and the local community information DB 234, and four functional units, the KPI analysis unit 222, the causal relationship management unit 224, the revision proposal unit 226, and the transportation plan management unit 228. Furthermore, each of the transportation service information DB 232 and the local community information DB 234 may include a subordinate database, and each of the KPI analysis unit 222, the causal relationship management unit 224, the revision proposal unit 226, and the transportation plan management unit 228 may include a subordinate functional unit. Therefore, the subordinate databases and subordinate functional units included in each database and functional unit will be described here.
[0043] The transportation service information DB 232 is a database for storing transportation service information related to transportation services, and as shown in Figure 3, may include an existing transportation plan information DB 401, a transportation service KPI information DB 402, an infrastructure information DB 403, a vehicle configuration information DB 404, a map information DB 405, a weather information DB 407, and a passenger movement information DB 406. As described above, the various information included in the transportation service information DB 232 may be obtained, for example, from a transportation company that provides the transportation service.
[0044] The existing transportation plan information DB401 is a database that stores existing transportation plan information that indicates existing transportation plans for the transportation service. This existing transportation plan information may include, for a predetermined period of time, information that identifies the transportation route to be implemented as the transportation service, information that identifies the vehicle to be used to implement the transportation route, information that indicates the departure and arrival locations of the transportation route, information on the number of transportation routes, information on the required time for the transportation route, and information on the number of passengers for the transportation route. As will be described later, this existing transportation plan information is used to generate revised transportation plan information that improves inconsistent KPIs, and may be created and input by, for example, a user (client, etc.) of the transportation plan management device.
[0045] The transportation service KPI information DB 402 stores information on KPIs related to existing transportation plan information stored in the existing transportation plan information DB 401 and information indicating required thresholds for each KPI. As described above, these KPIs are quantitative indicators that indicate the performance of transportation services with respect to various elements in the existing transportation plan information. For example, in one embodiment, these KPIs may include performance indicators for the operating costs, expected revenue, environmental impact, and energy efficiency of the transportation plan. Furthermore, the required thresholds are information that specify the required performance range for each KPI. In the present disclosure, a KPI that does not meet the required threshold (i.e., deviates from the range) is referred to as an "inconsistent KPI." As an example, if the existing transportation plan information includes a KPI such as "vehicle utilization rate = 45%" and the required threshold for that KPI is "60% to 70%," the transportation service KPI information DB 402 identifies the KPI as an "inconsistent KPI" that does not meet the required threshold.
[0046] The infrastructure information DB 403 is a database that stores infrastructure information indicating existing infrastructure in the area where the transportation service is provided. This existing infrastructure may include, for example, information identifying tracks, stations, vehicles, electric wires, ticket gates, etc. used in the transportation service.
[0047] The vehicle configuration information DB 404 is a database that stores information indicating the configuration of vehicles (trains, buses, etc.) used in the transportation service. This vehicle configuration may include, for example, information identifying the vehicles in the train, the number of passengers in each vehicle, the vehicles before and after a specific vehicle, etc.
[0048] The map information DB 405 is a database that stores map information related to the area where the transportation service is provided. This map information may include two-dimensional or three-dimensional maps, satellite photos, information related to the elevation of the area, etc.
[0049] Weather information 406 is a database that stores weather information related to the area where the transportation service is provided. This weather information may include information indicating the temperature, humidity, rainfall, snowfall, sunshine hours, etc., of the area where the transportation service is provided.
[0050] The passenger movement information DB 407 is a database that stores passenger movement information indicating the movement history of passengers who use the transportation service. This passenger movement information may include, for example, information such as the date and time, type (train or bus), departure location, arrival location, required time, and waiting time of the transportation route used by the passenger in the past.
[0051] The local community information DB 234 is a database for storing local community information related to the local community where the transportation service is provided, and as shown in Figure 3, may include a population information DB 411, a resident movement information DB 412, a transportation means information DB 413, a tax information DB 414, an economic information DB 415, and a customer satisfaction information DB 416. As described above, the various information stored in this local community information DB 234 may be obtained from any third party such as a local government, administrative agency, company, or organization, or may be obtained by sensors installed in the community.
[0052] The population information DB 411 is a database that stores population information indicating the population of the area where the transportation service is provided. This population information may include, for example, information indicating the population living in the area and the floating population staying in the area for work, school, tourism, etc.
[0053] The resident movement information DB 412 is a database that stores resident movement information indicating the movement of residents living in the area where the transportation service is provided. This resident movement information may include, for example, information on the number of people who moved into the area and information on the number of people who moved out of the area over a predetermined period of time.
[0054] The transportation information DB 413 is a database that stores transportation information indicating transportation methods provided in a specific area. This transportation information may include, for example, information about services such as buses, trains, and monorails provided in the specific area and their usage records.
[0055] The tax information DB 414 is a database that stores tax information related to residents living in the area where the transportation service is provided. This tax information may include, for example, information related to the average tax amount, income, and delinquent tax of residents living in the area.
[0056] The economic information DB 415 is a database that stores economic information related to the economy of the region where the transportation service is provided. This economic information may include information on the major industries, production information, employment rate, etc. in the region.
[0057] The customer satisfaction information DB 416 is a database that stores customer satisfaction information indicating the satisfaction level of customers of the transportation service who reside in the area where the transportation service is provided. This customer satisfaction information may include information regarding customer satisfaction with multiple transportation services and transportation itineraries, and customer travel preferences (preferring routes with shorter travel times, preferring routes with fewer transfers, etc.).
[0058] Of the various types of information stored in the transportation service information DB 232 and the local community information DB 234, the existing transportation plan information DB 401 and the transportation service KPI information DB 402 are used to identify inconsistent KPIs that do not meet the required thresholds in the existing transportation plan information, as described below. On the other hand, the various types of information stored in the infrastructure information DB 403, the vehicle configuration information DB 404, the map information DB 405, the weather information DB 406, the passenger movement information DB 407, the population information DB 411, the resident movement information DB 412, the transportation means information DB 413, the tax information DB 414, the economic information DB 415, and the customer satisfaction information DB 416 are used to identify causal relationships related to the identified inconsistent KPIs. In the following, the information stored in the infrastructure information DB 403, vehicle configuration information DB 404, map information DB 405, weather information DB 406, passenger movement information DB 407, population information DB 411, resident movement information DB 412, transportation means information DB 413, tax information DB 414, economic information DB 415 and customer satisfaction information DB 416 and used to identify causal relationships will be collectively referred to as "statistical information" 315.
[0059] The KPI analysis unit 222 includes an inconsistent KPI identification unit 421 that identifies inconsistent KPIs that do not satisfy a predetermined required threshold based on the existing transportation plan information and transportation service KPI information stored in the transportation service information DB 232, and an information extraction unit 422 that extracts information (a subset of statistical information) for determining a causal relationship related to the identified inconsistent KPI from the statistical information 315. Note that details of the operation of the KPI analysis unit 222 will be described later, and therefore will not be described here.
[0060] The causal relationship management unit 224 includes a causal relationship determination unit 431 that determines the causal relationship that exists between the inconsistent KPI identified by the KPI analysis unit 222 and the related variables in the extracted subset of statistical information, and a hypothesis inference unit 432 that generates inconsistency hypothesis information that indicates the reason why the inconsistent KPI does not satisfy the required threshold. Note that the details of the operation of the causal relationship management unit 224 will be described later, so the description thereof will be omitted here.
[0061] The revision proposal unit 226 includes an operation determination unit 441 that generates variable change operation information indicating a recommended variable change operation for a related variable that is in a causal relationship with the inconsistency KPI, based on the inconsistency hypothesis information determined by the causal relationship management unit 224, and a KPI change calculation unit 442 that calculates the amount of change in the inconsistency KPI when the variable change operation is performed. Note that the details of the operation by the revision proposal unit 226 will be described later, and therefore will not be described here.
[0062] The transportation plan management unit 228 includes a reception unit 451 that outputs a causal relationship diagram showing the determined causal relationship, inconsistency hypothesis information, variable change operation information, and the amount of change in the inconsistent KPI when the variable change operation is performed, and receives input from the user, and a transportation plan generation unit 452 that, when an approval input approving the variable change operation is received, reflects the variable change operation in the existing transportation plan information, thereby generating revised transportation plan information in which the inconsistent KPI is revised to satisfy the required threshold.
[0063] The transportation plan management device 210 shown in Figure 3 can analyze the reasons why a KPI does not meet a required threshold while taking into account the causal relationships between numerous variables and KPIs for transportation plans for transportation services, and then generate a revised transportation plan that improves the KPI.
[0064] Next, with reference to FIG. 4, an overall flow of a transportation plan management method according to an embodiment of the present disclosure will be described.
[0065] 4 is a diagram showing an example of the overall flow of a transportation plan management method 400 according to an embodiment of the present disclosure. The transportation plan management method 400 shown in Fig. 4 is a method for generating a revised transportation plan in which the KPI is improved after deriving the reason why a KPI that does not satisfy a required threshold does not satisfy the required threshold while taking into account the causal relationships between numerous variables and KPIs regarding a transportation plan for a transportation service, and is implemented by the functional units of the transportation plan management device 210 shown in Figs. 2 and 3.
[0066] First, in step S402, the KPI analysis unit 222 acquires the transportation service information stored in the transportation service information DB 232 and the local community information stored in the local community information DB 234. As described above, the acquired information includes existing transportation plan information and transportation service KPI information used to identify inconsistent KPIs, and statistical information (e.g., statistical information 315 shown in FIG. 3 ) used to identify causal relationships. Here, the case where the KPI analysis unit 222 acquires all of the transportation service information and local community information has been described as an example. However, the present disclosure is not limited to this. For example, if the transportation service information and local community information are stored on a remote third-party server, the KPI analysis unit 222 can first acquire the existing transportation plan information and transportation service KPI information used to identify inconsistent KPIs, and then, after identifying the inconsistent KPIs, acquire only the statistical information related to these inconsistent KPIs from the server. This saves bandwidth for transmitting information that is less relevant to the inconsistent KPIs.
[0067] Next, in step S404, the inconsistent KPI identification unit 421 of the KPI analysis unit 222 identifies inconsistent KPIs based on the existing transportation plan information and the transportation service KPI information acquired in step S402. As described above, the transportation service KPI information is information indicating KPIs related to the existing transportation plan information and the required thresholds for each KPI. Therefore, the inconsistent KPI identification unit 421 can identify KPIs that do not meet the specified required thresholds in the transportation service KPI information corresponding to specific existing transportation plan information as inconsistent KPIs. As an example, if the transportation service KPI information specifies a required threshold of "60% to 70%" for the KPI "vehicle utilization rate," and the existing transportation plan information specifies "vehicle utilization rate = 45%, the inconsistent KPI identification unit 421 may identify "vehicle utilization rate" as an inconsistent KPI.
[0068] Next, in step S406, the information extraction unit 422 of the KPI analysis unit 222 receives an input from the user specifying a dimension parameter that defines the search range of statistical information based on the influence range of the inconsistent KPI. The influence range of the inconsistent KPI here refers to information for identifying where, when, and what the inconsistent KPI affects. The dimension parameter is set based on the influence range of the inconsistent KPI and defines the maximum search range for extracting information related to the inconsistent KPI (a subset of statistical information) from the statistical information. In one embodiment, the user of the transportation plan management device 210 may set the dimension parameter in a dimension parameter setting interface 1200 shown in FIG. 12 and displayed via the user terminal 260. In one embodiment, the dimension parameters may include, for a search performed on statistical information, a time parameter that specifies a temporal range (e.g., from what time until what time the inconsistent KPI occurred), a location parameter that specifies a spatial range (e.g., the geographical location or coordinates where the inconsistent KPI occurred), a vehicle parameter that specifies a target object range (e.g., from how many cars in front of the target vehicle the inconsistent KPI occurred until how many cars behind the target vehicle), a station parameter that specifies a station range (e.g., how many stations in front of the target station the inconsistent KPI occurred until how many stations behind the target station the inconsistent KPI occurred), and characteristic information that specifies the target object range (e.g., the energy efficiency of the vehicle in which the inconsistent KPI occurred). Note that the dimension parameters may be set to include the same range as the range of influence of the inconsistent KPI; however, because related variables that have a causal relationship with the inconsistent KPI may exist outside the range of influence of the inconsistent KPI, it is desirable to set the dimension parameters to a range wider than the range of influence of the inconsistent KPI. Details of steps S406 to S412 will be described later in the causal relationship determination process 500 with reference to FIG. 5, and therefore will not be described here.
[0069] Next, in step S408, the information extraction unit 422 extracts a subset of statistical information for determining the causal relationship regarding the inconsistent KPI from the statistical information acquired in step S402 (statistical information 315 shown in Figure 3) based on the dimension parameters set in step S406.
[0070] Next, in step S410, the causal relationship determination unit 431 of the causal relationship management unit 224 determines a causal relationship regarding the inconsistent KPI using the subset of statistical information extracted in step S408. Here, the causal relationship determination unit 431 may determine multiple causal relationships for one inconsistent KPI. This causal relationship is a causal relationship that exists between the inconsistent KPI and any data element in the statistical information (hereinafter referred to as a "related variable"), and includes a relationship in which the inconsistent KPI is affected by the related variable and a relationship in which the inconsistent KPI affects the related variable.
[0071] Next, in step S412, the causal relationship determination unit 431 of the causal relationship management unit 224 calculates accuracy for each causal relationship determined in step S410. This accuracy is a value indicating the accuracy of a specific causal relationship, and may be a credibility calculated when determining a causal relationship.
[0072] Next, in step S414, the causal relationship determination unit 431 identifies the causal relationship with the highest accuracy for each inconsistent KPI. This is used to derive the reason why the inconsistent KPI occurred as the most certain causal relationship that affects (or is affected by) the specific inconsistent KPI. In one embodiment, if there are multiple causal relationships that satisfy a predetermined accuracy threshold, these multiple causal relationships may be identified.
[0073] Next, in step S416, the causal relationship determination unit 431 determines whether processing has been completed for all inconsistent KPIs identified in step S404 (whether the most accurate causal relationship has been identified). If processing has not been completed for all inconsistent KPIs, the process returns to step S406, and subsequent steps are performed for the target inconsistent KPI. On the other hand, if processing has been completed for all inconsistent KPIs, the process proceeds to step S418.
[0074] Next, in step S418, the hypothesis reasoning unit 432 of the causal relationship management unit 224 generates inconsistency hypothesis information indicating the reason why the inconsistent KPI does not satisfy the required threshold, based on the causal relationship with the highest accuracy identified in step S414. Note that the details of steps S418 to S424 will be described later in the causal relationship analysis process 600 shown in Figure 6 and the transportation plan generation process 700 shown in Figure 7, and therefore will not be described here.
[0075] Next, in step S420, the action determination unit 441 of the revision proposal unit 226 generates variable change action information indicating a variable change action for making the inconsistent KPI satisfy the required threshold based on the inconsistency hypothesis information generated in step S418, and the KPI change calculation unit 442 calculates the amount of change in the inconsistent KPI when the variable change action is performed. The variable change action here is information that specifies a recommended change to a variable that has a causal relationship with the inconsistent KPI. As an example, if it is determined that the inconsistent KPI with "congestion level" has a causal relationship with the related variable with "number of trains," the action determination unit 441 may determine a variable change action that recommends "adjusting the number of trains."
[0076] Next, in step S422, the reception unit 451 of the transportation plan management unit 228 outputs a causal relationship diagram showing the causal relationship between the inconsistent KPI and the related variables, inconsistency hypothesis information, variable change operation information, and the amount of change in the inconsistent KPI when the variable change operation is performed, and displays these to the user of the transportation plan management device 210, and receives feedback indicating whether or not to approve the variable change operation.
[0077] Next, in step S424, if the transport plan generation unit 452 of the transport plan management unit 228 receives approval input from the user approving the variable change operation output in step S422, the transport plan generation unit 452 generates revised transport plan information by reflecting the variable change operation in the existing transport plan information. This revised transport plan information is transport plan information in which related variables have been adjusted so that the inconsistent KPI satisfies the required threshold.
[0078] 4, the transportation plan management method 400 determines the causal relationship between an inconsistent KPI that does not satisfy a predetermined required threshold and a related variable in statistical information related to the inconsistent KPI, and based on the causal relationship, it is possible to derive the reason why the inconsistent KPI did not satisfy the required threshold. Furthermore, based on the reason derived in this way, it is possible to generate revised transportation plan information in which the related variable is adjusted so that the inconsistent KPI satisfies the required threshold.
[0079] Next, the flow of the causal relationship determination process according to the embodiment of the present disclosure will be described with reference to FIG.
[0080] 5 is a diagram illustrating an example of the flow of a causal relationship determination process 500 according to an embodiment of the present disclosure. The causal relationship determination process 500 illustrated in FIG. 5 is a process for extracting a subset of statistical information related to the inconsistent KPI from the statistical information based on dimension parameters that define a search range for the statistical information, and determining a causal relationship that exists between the inconsistent KPI and related variables in the extracted subset of statistical information. The causal relationship determination process 500 is performed by the information extraction unit 422 of the KPI analysis unit 222 and the causal relationship determination unit 431 of the causal relationship management unit 224. Note that FIG. 5 illustrates an example in which the influence range of the inconsistent KPI is defined using parameters related to time, vehicles, and stations. However, the present disclosure is not limited thereto, and the influence range of the inconsistent KPI may be defined using any other parameters, such as the energy efficiency of the vehicle.
[0081] First, in step S502, the information extraction unit 422 determines the range of influence of the inconsistent KPI. Here, the information extraction unit 422 may determine the range of time, vehicles, and stations that deviate from the required threshold for the inconsistent KPI identified in step S404 of the above-described transportation plan management method 400, for example, based on the existing transportation plan information stored in the existing transportation plan information DB 401 and the KPI information stored in the transportation service KPI information DB 402. As an example, the information extraction unit 422 may determine the start time, starting vehicle, starting station, end time, ending vehicle, ending station, etc. of the inconsistent KPI as the range of influence of the inconsistent KPI.
[0082] Next, in step S504, the information extraction unit 422 sets the dimension parameters and step size based on the range of influence of the inconsistent KPI determined in step S504. In some embodiments, the information extraction unit 422 may display the range of influence of the inconsistent KPI determined in step S504 to the user, for example, via the dimension parameter setting interface 1200 described below, and allow the user to adjust and confirm the values of the dimension parameters and step size, and then finalize and set them. The step size here refers to the size by which target elements (l, m, and n described below) are expanded at one time when searching for statistical information using the dimension parameters, and may be set for each dimension parameter. In principle, the smaller the step size, the more detailed the subset of statistical information extracted, enabling more accurate causal relationship determination. However, this increases processing time. Therefore, it is desirable to set this step size taking into account the desired accuracy of causal relationship determination and processing time.
[0083] As an example, the information extraction unit 422 may set dimension parameters for searching statistical information based on the influence range of the inconsistent KPI, such as a time parameter L (e.g., 60 minutes) that defines a time range and its step size a (e.g., 3 minutes), a vehicle parameter M (e.g., 10 vehicles before and after) that defines a vehicle range and its step size b (e.g., 1 vehicle), and a station parameter N (e.g., 6 stations) that defines a station range and its step size c (e.g., 1 station). These dimension parameters define the maximum range of the search performed on the statistical information. The dimension parameters may be set to include the same range as the influence range of the inconsistent KPI. However, because related variables that have a causal relationship with the inconsistent KPI may exist outside the influence range of the inconsistent KPI, it is desirable to set the time parameter L, the vehicle parameter M, and the station parameter N to a range wider than the influence range of the inconsistent KPI.
[0084] Next, in step S506, the information extraction unit 422 initializes the value of a target element for searching for statistical information and starts searching for statistical information. The target element here is a value that identifies specific data in the statistical information and may be set based on a dimension parameter. For example, if the dimension parameters include a time parameter L, a vehicle parameter M, and a station parameter N, the information extraction unit 422 may set a target time l that identifies a specific time, a target vehicle m that identifies a specific vehicle, and a target station n that identifies a specific station. Then, the information extraction unit 422 sets the target time l, the target vehicle m, and the target station n to "0" and determines whether information corresponding to the target time l, the target vehicle m, and the target station n exists in the statistical information.
[0085] Next, in step S508, the information extraction unit 422 identifies information corresponding to any of the target time 1, target vehicle m, and target station n from the statistical information, and extracts it as a subset of the statistical information. More specifically, the information extraction unit 422 may extract information that satisfies a search range defined in the following mathematical formula 1 as the subset of the statistical information. As an example, the information extraction unit 422 may acquire, from the statistical information, passenger movement information associated with timestamps within the time range (l1, l2) from the passenger movement information DB 407, acquire vehicle formation information of vehicles within the vehicle range (m1, m2) from the vehicle formation information DB 404, and acquire map information present in an area within the station range (n1, n2) from the map information DB 405. In this way, information that is included in the search range set based on the influence range of the inconsistent KPI and that may have a causal relationship with the inconsistent KPI can be extracted from the statistical information as a subset of the statistical information.
[0086] Next, in step S510, the causal relationship management unit 224 analyzes the subset of statistical information extracted in step S508 and the inconsistent KPI identified in step S404 of the transportation plan management method 400 described above to determine a causal relationship regarding the inconsistent KPI. The causal relationship here refers to a causal relationship between the inconsistent KPI and a related variable in the statistical information, including a relationship in which the inconsistent KPI is affected by a related variable and a relationship in which the inconsistent KPI affects a related variable. The related variable here may be data related to any feature or attribute in the statistical information. As an example, the causal relationship management unit 224 may determine that the inconsistent KPI with "congestion level" and the related variable with "number of trains" are in a causal relationship. Here, a machine learning model trained to determine causal relationships may be used as a means for analyzing the subset of statistical information and the inconsistent KPI and determining a causal relationship, but is not particularly limited thereto. In addition, when determining the causal relationship, the causal relationship management unit 224 may generate related variables that are causally related to the target inconsistent KPI and a causal relationship formula that indicates the degree of influence of each related variable on the inconsistent KPI.
[0087] Next, in step S512, the causal relationship management unit 224 calculates the accuracy of the causal relationship of the inconsistent KPI determined in step S510. This accuracy is a value indicating the accuracy of a specific causal relationship, and may be a credibility score calculated when determining the causal relationship. In an embodiment, the causal relationship management unit 224 may calculate the accuracy of the causal relationship using a causal relationship formula generated when determining the causal relationship.
[0088] Next, in step S514, the causal relationship management unit 224 determines whether the target time l is equal to or greater than the time parameter L. If the target time l is equal to or greater than the time parameter L, the process proceeds to step S518. If the target time l is not equal to or greater than the time parameter L, the process proceeds to step S516.
[0089] Next, in step S516, the causal relationship management unit 224 adds the step size a of the target time l to the target time l, and returns to step S508.
[0090] Next, in step S518, the causal relationship management unit 224 determines whether the target vehicle m is equal to or greater than the vehicle parameter M. If the target vehicle m is equal to or greater than the vehicle parameter M, the process proceeds to step S522. If the target vehicle m is not equal to or greater than the vehicle parameter M, the process proceeds to step S520.
[0091] Next, in step S520, the causal relationship management unit 224 adds the step size b of the target vehicle m to the target vehicle m, and returns to step S508.
[0092] Next, in step S522, the causal relationship management unit 224 determines whether the target station n is equal to or greater than the station parameter N. If the target station n is equal to or greater than the station parameter N, the process proceeds to step S524. If the target station n is not equal to or greater than the station parameter N, the process ends.
[0093] Next, in step S524, the causal relationship management unit 224 adds the step size c of the target station n to the target station n, and returns to step S508.
[0094] 5, the causal relationship determination process 500 uses dimension parameters to extract information from statistical information that corresponds to a search range including the time, location, and target at which the inconsistent KPI occurred and that may have a causal relationship with the inconsistent KPI. The extracted information can then be analyzed together with the inconsistent KPI to determine the causal relationship related to the inconsistent KPI. As described below, the causal relationship derived in this manner can be used to determine why a specific KPI did not meet the required threshold, or to generate revised transportation plan information that has been revised so that the inconsistent KPI meets the required threshold.
[0095] Next, the flow of a causal relationship analysis process 600 according to an embodiment of the present disclosure will be described with reference to FIG.
[0096] 6 is a diagram illustrating an example of the flow of a causal relationship analysis process 600 according to an embodiment of the present disclosure. The causal relationship analysis process 600 illustrated in FIG. 6 is a process for analyzing the causal relationship of the inconsistent KPI determined by the above-described causal relationship determination unit 431, generating inconsistency hypothesis information indicating the reason why the inconsistent KPI does not satisfy the required threshold, and generating recommended variable change actions so that the inconsistent KPI satisfies the required threshold, and is performed by the causal relationship management unit 224 and the revision proposal unit 226.
[0097] First, in step S602, the hypothesis reasoning unit 432 of the causal relationship management unit 224 acquires information on the causal relationship that exists between the inconsistent KPI and the related variables, as determined by the above-mentioned causal relationship determination unit 431 (see the transportation plan management method 400 and the causal relationship determination process 500).
[0098] Next, in step S604, the hypothesis reasoning unit 432 sets an accuracy threshold for evaluating the causal relationships. This accuracy threshold is information that defines a boundary line for distinguishing, from among the determined causal relationships, those causal relationships that are highly certain and accurate enough to be used to create inconsistent hypothesis information from those that are not, and may be, for example, a lower limit of the accuracy value. This accuracy threshold may be freely set by, for example, a user of the transportation plan management device 210. In one embodiment, this accuracy threshold may be, for example, "70%."
[0099] Next, in step S606, the hypothetical reasoning unit 432 determines whether the causal relationship of each inconsistent KPI acquired in step S602 satisfies the accuracy threshold set in step S604. If the accuracy of the causal relationship of a specific inconsistent KPI satisfies the accuracy threshold, the process proceeds to step S608, and if the accuracy of the causal relationship of the inconsistent KPI does not satisfy the accuracy threshold, the process proceeds to step S618.
[0100] In step S618, if the accuracy of the causal relationship of the inconsistent KPI does not meet the accuracy threshold, the hypothesis reasoning unit 432 determines that the inconsistent KPI is independent (i.e., not influenced by other variables) because it was not possible to determine a reliable causal relationship for the inconsistent KPI, and proceeds to step S620.
[0101] In step S608, the hypothetical reasoning unit 432 generates an element list related to the causal relationships of the inconsistent KPIs that satisfy the accuracy threshold. This element list is a list that lists all KPIs and related variables that are in a causal relationship, and may be generated, for example, based on terms in the causal relationship formula generated when determining the causal relationship. As an example, if the causal relationship formula is Formula 2 shown below, the element list would be "Data1, Data2, Data4."
[0102] Next, in step S610, the abductive reasoning unit 432 calculates the degree of influence of each element in the element list generated in step S608 on the inconsistent KPI. The influence here is a quantitative measure of the degree of influence that a specific element has on the inconsistent KPI, and can be calculated using the causal relationship formula generated when determining the causal relationship. In some embodiments, the abductive reasoning unit 432 may determine a database (e.g., the weather information DB 406, the economic information DB 415, etc. shown in FIG. 3 ) in which each element in the element list is identified.
[0103] Next, in step S612, the hypothetical reasoning unit 432 identifies key variables that satisfy a predetermined influence criterion based on the influence of each element calculated in step S610. This influence criterion is information that defines a boundary line for distinguishing between variables that have a significant influence on the inconsistent KPI and variables that do not, and may be, for example, a lower limit of the influence. This influence criterion may be freely set by, for example, a user of the transportation plan management device 210. Furthermore, the key variables are variables that have a significant influence from among variables that have a causal relationship with the inconsistent KPI.
[0104] Next, in step S614, the abductive reasoning unit 432 generates inconsistency hypothesis information indicating the reason why the inconsistent KPI did not satisfy the required threshold based on the primary variable identified in step S610. Here, the reason why the inconsistent KPI did not satisfy the required threshold may be expressed in natural language. Here, the abductive reasoning unit 432 may generate the inconsistency hypothesis information according to the polarity (direct or inverse proportionality) of the relationship between the primary variable and the inconsistent KPI in the causal relationship formula generated when determining the causal relationship. For example, if KPI1 is below the required threshold and primary variable A is directly proportional to KPI1, the abductive reasoning unit 432 may generate, as the inconsistency hypothesis information, a sentence indicating that "KPI1 was determined to be inconsistent because primary variable A was low." As another example, if KPI1 is below the required threshold and primary variable A is inversely proportional to KPI1, the abductive reasoning unit 432 may generate, as the inconsistency hypothesis information, a sentence indicating that "KPI1 was determined to be inconsistent because primary variable A was high." In some embodiments, the abductive reasoning unit 432 may use an existing abductive reasoning method (e.g., a machine learning model trained to perform abductive reasoning) to determine inconsistent hypothesis information.
[0105] Next, in step S616, the action determination unit 441 of the revision proposal unit 226 generates variable change action information indicating a variable change action for making the inconsistency KPI satisfy the required threshold based on the inconsistency hypothesis information generated in step S414, and the KPI change calculation unit 442 calculates the amount of change in the inconsistency KPI when the variable change action is performed. Here, if there are multiple variable change actions for making the inconsistency KPI satisfy the required threshold, the action determination unit 441 may generate these multiple variable change actions as candidates to be presented to the user.
[0106] As described above, the variable change operation here is information specifying a recommended change to a variable that has a causal relationship with the inconsistency KPI, and can be determined based on the inconsistency hypothesis information generated in step S414. In one embodiment, the operation determination unit 441 may determine how the primary variable needs to be changed so that the inconsistency KPI satisfies the required threshold by using the causal relationship formula generated during the causal relationship determination and the inconsistency hypothesis information. As an example, if the inconsistency KPI with "congestion level" is determined to have a causal relationship with the primary variable with "number of trains," and inconsistency hypothesis information indicating the reason that "the congestion level is determined not to satisfy the threshold because the number of trains is low" is determined, the operation determination unit 441 may determine a variable change operation indicating a recommended change to "number of trains" so that the inconsistency KPI with "congestion level" satisfies the required threshold, based on the inconsistency hypothesis information and the causal relationship formula. Furthermore, after the variable change operation is determined, the KPI change calculation unit 442 calculates the amount of change in the inconsistent KPI, which indicates how the inconsistent KPI will change when the variable change operation is performed. As an example, the KPI change calculation unit 442 may calculate that when the variable change operation of "increase the number of trains per hour by one" is performed, the KPI with "operating cost" will be "+10%" and the inconsistent KPI with "congestion level" will be "-25%".
[0107] Next, in step S620, the operation determination unit 441 determines whether processing has been completed for all causal relationships determined in step S606 to satisfy the accuracy threshold (whether inconsistency hypothesis information, variable change operation information, and KPI change amount information have been generated). If processing has not been completed for all causal relationships, the process returns to step S606, and subsequent steps are performed for unprocessed causal relationships. On the other hand, if processing has been completed for all causal relationships, the process proceeds to the revised transportation plan generation process 700 shown in FIG. 7.
[0108] According to the causal relationship analysis process 600 shown in FIG. 6, by analyzing the causal relationships determined for an inconsistent KPI that does not satisfy a predetermined required threshold, it is possible to generate inconsistency hypothesis information indicating the reason why the inconsistent KPI did not satisfy the required threshold, and to generate variable change operation information indicating appropriate changes to key variables that have a significant impact on the inconsistent KPI so that the inconsistent KPI satisfies the required threshold.
[0109] Next, a transportation plan generation process according to an embodiment of the present disclosure will be described with reference to Fig. 7. Fig. 7 is a diagram showing an example of the flow of a transportation plan generation process 700 according to an embodiment of the present disclosure. The transportation plan generation process 700 shown in Fig. 7 is a process for generating a revised transportation plan based on the inconsistency hypothesis information and variable change operations generated in the causal relationship analysis process 600, and is performed by the transportation plan management unit 228 shown in Fig. 2.
[0110] First, in step S702, the reception unit 451 of the transportation plan management unit 228 acquires information on inconsistent KPIs, causal relationship diagrams, inconsistency hypothesis information, variable change operations, and KPI change amounts from the KPI analysis unit 22, the causal relationship management unit 224, and the revision proposal unit 226.
[0111] Next, in step S704, the reception unit 451 checks whether the transportation plan is set to be automatically generated. If the transportation plan is not set to be automatically generated, the process proceeds to step S706. If the transportation plan is set to be automatically generated, the process proceeds to step S714.
[0112] Next, in step S706, the reception unit 451 outputs the inconsistent KPI, causal diagram, inconsistency hypothesis information, variable change operation, and KPI change amount information to the user terminal 260, for example, via the input / output unit 246 shown in Figure 2, and displays it to the user.
[0113] Next, in step S708, the accepting unit 451 accepts feedback on the information displayed on the user terminal 260 in step S706 from the user, for example, via the input / output unit 246 shown in Fig. 2. This feedback may include, for example, an input approving the variable change operation displayed in step S706 or an input specifying a modification of the variable change operation.
[0114] Next, in step S710, the reception unit 451 determines whether the variable change operation has been approved based on the feedback received in step S708. If the variable change operation has not been approved, the process proceeds to step S712. If the variable change operation has been approved, the process proceeds to step S714.
[0115] If the variable change operation is not approved, in step S712, the operation determination unit 441 of the revision proposal unit 226 generates a new variable change operation to make the inconsistent KPI satisfy the required threshold based on the user's input to modify the variable change operation and the inconsistency hypothesis information generated for the inconsistent KPI, and the KPI change calculation unit 442 calculates the amount of change in the inconsistent KPI when this new variable change operation is performed. After that, the process returns to step S706.
[0116] If the variable change operation is approved, in step S714, the transportation plan generation unit 452 generates a revised transportation plan by reflecting the approved variable change operation in the existing transportation plan information, thereby revising the inconsistent KPIs that did not satisfy a predetermined required threshold in the existing transportation plan so that they satisfy the required threshold. As an example, if inconsistency hypothesis information is determined for the inconsistent KPI with the "congestion level" that indicates the reason that "the congestion level was determined not to satisfy the threshold because the number of trains is low," the operation determination unit 441 may generate a revised transportation plan by applying a variable change operation to the existing transportation plan information to appropriately increase the "number of trains" based on this inconsistency hypothesis information and the causal relationship formula so that the inconsistent KPI with the "congestion level" satisfies the required threshold.
[0117] Next, in step S716, the transportation plan generation unit 452 outputs the revised transportation plan generated in step S714 to the user terminal 260 via, for example, the input / output unit 246 shown in FIG. 2, and displays it.
[0118] According to the transportation plan generation process 700 described with reference to Figure 7, the reason why inconsistent KPIs that do not meet the required threshold do not meet the required threshold can be derived, and a revised transportation plan can be generated that reflects appropriate variable change operations to improve these KPIs in the existing transportation plan, and provided to the user.
[0119] Next, the concept of dimension parameters according to an embodiment of the present disclosure will be described with reference to FIGS. 8 to 10 . As described above, the dimension parameters according to an embodiment of the present disclosure are set based on the range of influence of an inconsistent KPI and are information that specifies the maximum range of a search for extracting information related to the inconsistent KPI (a subset of statistical information) from the statistical information. In one embodiment, the dimension parameters may include, for a search performed on the statistical information, a time parameter that specifies a temporal range, a location parameter that specifies a spatial range, a vehicle parameter that specifies a target range, a station parameter that specifies a station range, and feature information that specifies the target range. By using such dimension parameters, it is possible to identify information (a subset of statistical information) from the statistical information that is highly likely to have a causal relationship with a specific inconsistent KPI. Note that the dimension parameters may be set to include the same range as the range of influence of the inconsistent KPI. However, because related variables that have a causal relationship with the inconsistent KPI may exist outside the range of influence of the inconsistent KPI, it is desirable to set the dimension parameters to a range wider than the range of influence of the inconsistent KPI.
[0120] In Figures 8 to 10, the dimension parameters are explained with reference to graphs in which the horizontal axis represents time and the vertical axis represents stations, but this is an example to explain the concept of dimension parameters, and the types and application forms of dimension parameters are not limited to this.
[0121] FIG. 8 is a diagram illustrating an example of setting a time parameter L as a dimension parameter according to an embodiment of the present disclosure. In FIG. 8, the time parameter "L" represents the maximum time range (e.g., 60 minutes) of a search for extracting information related to an inconsistent KPI (a subset of statistical information) from statistical information. "L1" represents the time when an inconsistency in a specific KPI occurred, and "L2" represents the time when the inconsistency in the KPI was resolved. "x5" represents the time range corresponding to the inconsistent KPI (i.e., the time span from L1 to L2). "M" represents the maximum range of vehicles to be searched, and "N" represents the maximum range of stations to be searched. "t" represents the step size (e.g., 5 minutes) for searching the time range defined by the time parameter L. The target time "l" identifies a specific time when performing a search within the range of the time parameter L and is an integer incremented by the step size t within the range from 0 to L.
[0122] In this way, by setting the time parameter L as the dimension parameter, it is possible to identify and extract information (a subset of the statistical information) corresponding to the time range in which the inconsistent KPI occurred from the statistical information. Furthermore, by using the information corresponding to the time range in which the inconsistent KPI occurred to determine the causal relationship regarding the inconsistent KPI, it is possible to determine the causal relationship regarding the inconsistent KPI with high accuracy.
[0123] FIG. 9 is a diagram illustrating an example of setting a vehicle parameter M as a dimension parameter according to an embodiment of the present disclosure. In FIG. 9 , the vehicle parameter "M" indicates the maximum range of search target vehicles (e.g., 10 vehicles before and after) for extracting information related to an inconsistent KPI (a subset of statistical information) from statistical information. "M1" indicates a vehicle in which an inconsistency in a specific KPI occurred, and "M2" indicates a vehicle in which the inconsistency in the KPI has been resolved. "x5" indicates the range of vehicles corresponding to the inconsistent KPI (i.e., the width from M1 to M2). When M1 = M2, this indicates that an inconsistent KPI occurred for only one vehicle. "t" indicates the step size (e.g., one vehicle) for searching the vehicle range defined by the vehicle parameter M. The target vehicle "m" is an integer that identifies a specific vehicle when searching the range of the vehicle parameter M and is incremented by the step size t within the range from 0 to M.
[0124] In this way, by setting the vehicle parameter M as a dimension parameter, it is possible to identify and extract information (a subset of the statistical information) corresponding to the range of vehicles in which an inconsistent KPI has occurred from the statistical information. Furthermore, by using the information corresponding to the vehicle in which an inconsistent KPI has occurred to determine the causal relationship regarding the inconsistent KPI, it is possible to determine the causal relationship regarding the inconsistent KPI with high accuracy.
[0125] FIG. 10 is a diagram illustrating an example of setting a station parameter N as a dimension parameter according to an embodiment of the present disclosure. In FIG. 10, the station parameter "N" indicates the maximum range of stations (e.g., 10 stations before and after) to be searched for, extracting information related to an inconsistent KPI (a subset of statistical information) from statistical information. "N1" indicates a station where a specific KPI inconsistency occurred, and "N2" indicates a station where the KPI inconsistency has been resolved. "x5" indicates the range of stations corresponding to the inconsistent KPI (i.e., the width from N1 to N2). When N1 = N2, this indicates that an inconsistent KPI occurred for only one station. "t" indicates the step size (e.g., one station) used to search the station range defined by the station parameter N. The target station "n" is an integer that identifies a specific station when searching the range of the station parameter N and is incremented by a step size t within the range from 0 to N.
[0126] In this way, by setting the station parameter N as a dimension parameter, it is possible to identify and extract information (a subset of the statistical information) corresponding to the range of stations where an inconsistent KPI has occurred from the statistical information. Furthermore, by using the information corresponding to the station where the inconsistent KPI has occurred to determine the causal relationship regarding the inconsistent KPI, it is possible to determine the causal relationship regarding the inconsistent KPI with high accuracy.
[0127] Next, a cause and effect diagram according to an embodiment of the present disclosure will be described with reference to FIG.
[0128] 11 is a diagram illustrating an example of a causal relationship diagram 1100 according to an embodiment of the present disclosure. As described above, the causal relationship determination unit 431 according to an embodiment of the present disclosure can determine a causal relationship between an inconsistent KPI that does not satisfy a required threshold and a related variable in a subset of statistical information. In an embodiment, the causal relationship determination unit 431 may visualize the determined causal relationship as a causal relationship diagram 1100 as shown in FIG. 11 and provide it to a user.
[0129] The causal relationship diagram 1100 shown in FIG. 11 is a causal loop diagram (CLD) that uses a directed graph network to represent the presence or absence of a causal relationship between an inconsistent KPI and a related variable. In the causal relationship diagram 1100, KPIs 1110 related to transportation services and related variables 1120 in a subset of statistical information can be represented as separate nodes. Furthermore, the causal relationship between the KPIs 1110 and the related variables 1120 can be represented as a bidirectional arrow. As described above, these KPIs 1110 are KPIs stored in the transportation service KPI information DB 402, such as operational costs, projected revenue, environmental impact, and energy efficiency of a transportation plan, and the related variables 1120 are related variables in a subset of statistical information extracted from the transportation service information DB 232 and the local community information DB 234.
[0130] As an example, in the causal relationship diagram 1100, C1 " 1111 is "KPI P2 ", "KPI S2 ", "Data B1 " and "Data S4 " is connected to "KPI C1 " 1111 is "KPI P2 " and "KPI S2 " and "Data B1 " and "Data S4 It can be said that there is a causal relationship between the related variables and KPI. C1 " 1111 is "Congestion level", "KPI P2 " is "Number of trains", "KPI S2 " corresponds to the KPI "passenger waiting time" and "Data B1 " is "Rainy weather", "Data S4 " may correspond to a variable related to "resident population". C1 The causal relationship regarding "1111" can be expressed by a causal relationship formula generated when the causal relationship determination unit 431 determines the causal relationship, as shown in Formula 3 below.
[0131] As shown in Figure 11, the causal relationships determined for inconsistent KPIs are visualized as a causal relationship diagram 1100 as shown in Figure 11 and provided to the user, allowing the user to easily understand other KPIs and related variables that affect a specific KPI.
[0132] Next, a dimension parameter setting interface according to an embodiment of the present disclosure will be described with reference to FIG.
[0133] 12 is a diagram illustrating a dimension parameter setting interface 1200 according to an embodiment of the present disclosure. As described above, in the transportation plan management means according to an embodiment of the present disclosure, information (a subset of statistical information) used to determine a causal relationship is identified based on dimension parameters that are set based on the range of influence of the inconsistent KPI. These dimension parameters may be set by a user, for example, through the dimension parameter setting interface 1200 illustrated in FIG.
[0134] As shown in FIG. 12, the dimension parameter setting interface 1200 may include a parameter name setting window 1205, a range direction setting window 1210, a maximum value setting window 1215, a step size setting window 1220, an add / delete button 1225, and an enter button 1230.
[0135] The parameter name setting window 1205 is an interface window for setting the name of a dimension parameter. In some embodiments, the user may select a specific dimension parameter from among pre-prepared dimension parameters. In some embodiments, the user may custom-set a new dimension parameter. As an example, the dimension parameters may include "time" that defines the temporal range of a search performed on statistical information, "number of trains" that defines the range of the target object (target train), "number of stations" that defines the spatial range, and "energy consumption per kilometer" as a feature for defining the range of the target object (target train).
[0136] The range direction setting window 1210 is an interface window for specifying whether the search range of a particular dimension parameter should include before, after, or both before and after the target element (target time, target station, target vehicle, etc.).
[0137] The maximum value setting window 1215 is an interface window for specifying the maximum value of the search range for a specific dimension parameter. For example, as shown in Fig. 12, the maximum value of the dimension parameter "time" may be set to "60 minutes," and the maximum value of the dimension parameter "number of trains" may be set to "20."
[0138] The step size setting window 1220 is an interface window for specifying the minimum step value of a specific dimension parameter. In principle, the smaller the step size, the more precisely the subset of statistical information is extracted, enabling more accurate causal relationship determination. However, since the processing time increases, it is desirable to set this step size taking into consideration the desired accuracy of causal relationship determination and the processing time.
[0139] The add / delete button 1225 is a button for adding a new dimension parameter or deleting an existing dimension parameter.
[0140] The confirm button 1230 is a button for confirming and saving all the information about the dimension parameters after they have been input.
[0141] As described above, the dimension parameter setting interface 1200 described with reference to Fig. 12 makes it possible to set dimension parameters that indicate the range of influence of an inconsistent KPI. Furthermore, by using the dimension parameters set in this way, it is possible to extract a subset of statistical information related to a specific inconsistent KPI from the statistical information, thereby enabling highly accurate determination of causal relationships.
[0142] As described above, in the transportation plan management means according to an embodiment of the present disclosure, when an inconsistent KPI that does not satisfy a predetermined required threshold is identified in an existing transportation plan for a transportation service, a causal relationship between the inconsistent KPI and a related variable that affects the inconsistent KPI is determined, and a recommended variable change action for the related variable is determined based on the determined causal relationship. Then, by reflecting the determined variable change action in the existing transportation plan, a revised transportation plan can be generated in which the inconsistent KPI is revised to satisfy the required threshold.
[0143] In the transportation plan management means according to an embodiment of the present disclosure, the causal relationship regarding the inconsistent KPI is determined based on statistical information including information about the transportation service (e.g., infrastructure, vehicle configuration, passenger movement) and information about the local community to which the transportation service is provided (e.g., population, resident movement, taxes, economy, customer satisfaction). In this way, by identifying related variables that affect a specific KPI from a wide range of information, it becomes possible to suggest to the user various variable change operations to improve a specific inconsistent KPI, thereby improving the flexibility of the transportation plan.
[0144] Furthermore, in the transportation planning management means according to an embodiment of the present disclosure, information (a subset of statistical information) used to determine the causal relationship related to the inconsistent KPI is extracted from the statistical information based on dimension parameters that define the search range (time range, spatial range, range of object) that takes into account the range of influence of the inconsistent KPI. In this way, by using information limited to the range related to a specific inconsistent KPI, rather than all statistical information, to determine the causal relationship, it is possible to reduce processing time, save computing resources, and perform more detailed and accurate causal relationship determination.
[0145] Furthermore, in the transportation plan management means according to an embodiment of the present disclosure, after determining the causal relationship regarding the inconsistent KPI, a causal relationship diagram visualizing the causal relationship between the inconsistent KPI and related variables, various variable change actions for improving the inconsistent KPI, and information on the amount of change in the KPI when the variable change actions are adopted are output to the user. As a result, the user who has confirmed this information can easily grasp the related variables that affect a specific inconsistent KPI, the countermeasures for improving the KPI, and the impact of each measure on the KPI, thereby improving various KPIs in the transportation plan, such as operating costs, expected revenue, environmental impact, and energy efficiency, and realizing the efficiency of transportation service operations.
[0146] In this way, according to the transportation plan management means of an embodiment of the present disclosure, it is possible to analyze the reasons why a KPI does not meet the required threshold while taking into account the causal relationships between numerous variables and KPIs in the transportation plan for a transportation service, and then generate a revised transportation plan that improves the KPI.
[0147] As described above, the transportation plan management means according to the embodiment of the present disclosure includes the following aspects.
[0148] a causal relationship determination unit that analyzes the inconsistent KPI and the statistical information and determines a causal relationship between the inconsistent KPI and a related variable in a subset of statistical information extracted from the statistical information; and a transport plan generation unit that uses the existing transport plan information and the determined causal relationship to generate revised transport plan information in which the inconsistent KPI satisfies the required threshold. (Aspect 1) A transport plan management device comprising: a processor, a memory, and a storage unit;
[0149] (Aspect 2) The transportation planning management device according to Aspect 1, further comprising: an information extraction unit that receives input of dimension parameters from a user that define a search range for the statistical information based on the influence range of the inconsistent KPI; and extracts, from the statistical information, a subset of the statistical information that corresponds to the search range defined by the dimension parameters.
[0150] (Aspect 3) The transportation planning management device according to Aspect 2, characterized in that the dimension parameters include any one of a time parameter that defines the time range of the inconsistent KPI, a location parameter that defines the spatial range of the inconsistent KPI, a vehicle parameter that defines the object range of the inconsistent KPI, and a feature parameter that defines the object range of the inconsistent KPI.
[0151] (Aspect 4) The transportation planning management device according to any one of Aspects 1 to 3, further comprising: a hypothesis inference unit that calculates the degree of influence on the inconsistent KPI for each of the related variables determined to have a causal relationship with the inconsistent KPI; identifies a main variable among the related variables whose calculated degree of influence satisfies a predetermined influence standard; and generates inconsistency hypothesis information indicating the reason why the inconsistent KPI does not satisfy the required threshold based on the identified main variable.
[0152] (Aspect 5) The transportation plan management device according to Aspect 4, further comprising: an operation determination unit that generates variable change operation information indicating a recommended variable change operation for the key variable based on the generated inconsistency hypothesis information so that the inconsistency KPI satisfies the required threshold; and a KPI change calculation unit that calculates the amount of change in the inconsistency KPI when the variable change operation is performed.
[0153] (Aspect 6) The transportation plan management device according to Aspect 5, wherein the transportation plan generation unit outputs a causal relationship diagram showing the causal relationship between the inconsistent KPI and the key variable, the inconsistency hypothesis information, the variable change operation information, and an amount of change in the inconsistent KPI when the variable change operation is performed, and when an approval input approving the variable change operation is received, generates the revised transportation plan information by reflecting the variable change operation in the existing transportation plan information.
[0154] (Aspect 7) The transportation planning management device according to Aspects 1 to 6, characterized in that the statistical information includes infrastructure information indicating existing infrastructure in the area where the transportation service is provided, map information relating to the area where the transportation service is provided, weather information relating to the area where the transportation service is provided, population information relating to the area where the transportation service is provided, resident movement information relating to the area where the transportation service is provided, tax information relating to the area where the transportation service is provided, vehicle formation information indicating the formation of vehicles used in the transportation service, passenger movement information indicating the movement history of passengers using the transportation service, and customer satisfaction information indicating customer satisfaction with the transportation service.
[0155] Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present invention.
[0156] 150: Transportation plan management application, 200: Transportation plan management system, 210: Transportation plan management device, 220: Memory, 222: KPI analysis unit, 224: Causal relationship management unit, 226: Revision proposal unit, 228: Transportation plan management unit, 232: Transportation service information DB, 234: Local community information DB, 244: Processor, 246: Input / output unit, 250: Communication network, 260: User terminal
Claims
1. A transportation plan management device comprising a processor, a memory, and a storage unit, wherein the storage unit includes: existing transportation plan information indicating an existing transportation plan for a specified transportation service; transportation service KPI information including KPIs indicating performance indicators for the existing transportation plan information; and statistical information related to the transportation service, and the memory includes processing instructions for causing the processor to function as: an inconsistent KPI identification unit that identifies inconsistent KPIs that do not satisfy a specified required threshold among the KPIs based on the existing transportation plan information and the transportation service KPI information; a causal relationship determination unit that analyzes the inconsistent KPIs and the statistical information and determines a causal relationship between the inconsistent KPIs and related variables in a subset of statistical information extracted from the statistical information; and a transportation plan generation unit that generates revised transportation plan information in which the inconsistent KPIs satisfy the required threshold using the existing transportation plan information and the determined causal relationship.
2. The transportation planning management device of claim 1, further comprising: an information extraction unit that receives input of dimension parameters from a user that specify a search range for the statistical information based on the range of influence of the inconsistent KPI; and extracts from the statistical information a subset of the statistical information that corresponds to the search range specified by the dimension parameters.
3. The transportation planning management device of claim 2, characterized in that the dimension parameters include any one of a time parameter specifying the temporal range of the inconsistent KPI, a location parameter specifying the spatial range of the inconsistent KPI, a vehicle parameter specifying the object range of the inconsistent KPI, and a feature parameter specifying the object range of the inconsistent KPI.
4. The transportation planning management device of claim 1, further comprising a hypothesis inference unit that calculates the degree of influence on the inconsistent KPI for each of the related variables determined to be in a causal relationship with the inconsistent KPI, identifies a main variable among the related variables whose calculated degree of influence satisfies a predetermined influence criterion, and generates inconsistency hypothesis information indicating the reason why the inconsistent KPI does not satisfy the required threshold based on the identified main variable.
5. The transportation planning management device of claim 4, further comprising an amendment proposal unit having: an operation determination unit that generates variable change operation information indicating a recommended variable change operation for the key variable based on the generated inconsistency hypothesis information so that the inconsistency KPI satisfies the required threshold; and a KPI change calculation unit that calculates the amount of change in the inconsistency KPI when the variable change operation is implemented.
6. The transportation plan management device of claim 5, wherein the transportation plan generation unit outputs a causal relationship diagram showing the causal relationship between the inconsistent KPI and the main variable, the inconsistency hypothesis information, the variable change operation information, and the amount of change in the inconsistent KPI when the variable change operation is performed, and when an approval input approving the variable change operation is received, generates the revised transportation plan information by reflecting the variable change operation in the existing transportation plan information.
7. The transportation planning management device of claim 1, characterized in that the statistical information includes infrastructure information indicating existing infrastructure in the area where the transportation service is provided, map information relating to the area where the transportation service is provided, weather information relating to the area where the transportation service is provided, population information relating to the area where the transportation service is provided, resident movement information relating to the area where the transportation service is provided, tax information relating to the area where the transportation service is provided, vehicle composition information indicating the composition of vehicles used in the transportation service, passenger movement information indicating the movement history of passengers using the transportation service, and customer satisfaction information indicating customer satisfaction with the transportation service.
8. A transportation plan management method implemented by computer software in a transportation plan management device, the transportation plan management device comprising: a processor; a memory for storing program instructions; and a storage unit, the storage unit including: existing transportation plan information indicating an existing transportation plan for a specified transportation service; transportation service KPI information including KPIs indicating performance indicators for the existing transportation plan information; and statistical information related to the transportation service, the program instructions stored in the memory including: a step of identifying, based on the existing transportation plan information and the transportation service KPI information, an inconsistent KPI that does not satisfy a specified required threshold among the KPIs; a step of setting dimension parameters including any one of a time parameter that specifies the time range of the inconsistent KPI, a position parameter that specifies the spatial range of the inconsistent KPI, and a feature parameter that specifies the target range of the inconsistent KPI; and a step of extracting a subset of statistical information corresponding to the range specified in the dimension parameter from the statistical information. a step of analyzing the inconsistency KPI and the subset of statistical information, and determining a causal relationship between the inconsistency KPI and related variables in the subset of statistical information; a step of calculating an influence on the inconsistency KPI for each of the related variables determined to be in a causal relationship with the inconsistency KPI; a step of identifying a main variable among the related variables, the calculated influence of which satisfies a predetermined influence criterion; a step of generating inconsistency hypothesis information indicating a reason why the inconsistency KPI does not satisfy the required threshold based on the identified main variable; a step of generating variable change operation information indicating a recommended variable change operation for the main variable so that the inconsistency KPI satisfies the required threshold based on the generated inconsistency hypothesis information; a step of calculating an amount of change in the inconsistency KPI when the variable change operation is performed; and a step of outputting to a user a causal relationship diagram indicating a causal relationship between the inconsistency KPI and the main variable, the inconsistency hypothesis information, the variable change operation information, and the amount of change in the inconsistency KPI when the variable change operation is performed.a step of generating revised transportation plan information in which the inconsistent KPI satisfies the required threshold by reflecting the variable change operation in the existing transportation plan information when an approval input approving the variable change operation is received, the step of generating revised transportation plan information in which the inconsistent KPI satisfies the required threshold when the variable change operation is reflected in the existing transportation plan information.
Citation Information
Patent Citations
Business improvement support apparatus and business improvement support method
JP2018156346A
