Information processing device, information processing method, and program
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2026-01-30
- Publication Date
- 2026-08-13
Smart Images
Figure JP2026003293_13082026_PF_FP_ABST
Abstract
Description
Information Processing Apparatus, Information Processing Method, and Program
[0001] The present disclosure relates to an information processing apparatus, an information processing method, and a program.
[0002] Among the ground handling operations performed at an airport, the operations performed at the apron (also called a ramp) are called ramp operation tasks. For example, the ramp operation tasks include the process for taking an aircraft off.
[0003] However, among the ground handling operations, especially the ramp operation tasks are prone to delays. Therefore, in recent years, in order to ensure that an aircraft departs on time, improving the efficiency of ramp operation tasks has become an issue, and technologies for solving this issue have also been proposed.
[0004] For example, Patent Document 1 discloses a technique for determining that rescheduling is necessary and executing rescheduling when the delay time of the work in the ramp operation task exceeds a predetermined threshold value. Here, rescheduling means re-creating the work schedule.
[0005] Japanese Unexamined Patent Application Publication No. 2021-086444
[0006] The technique disclosed in Patent Document 1 executes rescheduling when a delay occurs in the work of the ramp operation task. However, if the cause of the delay in the work of the ramp operation task is unknown, it is considered difficult to improve the work, and thus it is difficult to improve the efficiency of the ramp operation task.
[0007] Therefore, an object of the present disclosure is to provide an information processing apparatus, an information processing method, and a program capable of improving the efficiency of ramp operation tasks in view of the above-described problems.
[0008] An information processing device according to one embodiment includes: an analysis target identification means for identifying an analysis target among a plurality of processes in airport ramp operation work in which a difference of a predetermined time or more occurs between the time the process is planned and the time the process is actually performed; a collection means for collecting at least one of the speech recognition results and text information of voices exchanged between the on-site personnel at the site where the analysis target was performed and other people before and during the execution of the analysis target identified process; and a reason generation means for generating the reason why the difference occurred in the analysis target identified process based on the speech recognition results and at least one of the text information, and outputting the generated reason.
[0009] An information processing method according to one embodiment is an information processing method performed by an information processing device, which includes: identifying as an analysis target a process among a plurality of processes in airport ramp operation work a process in which a difference of a predetermined time or more occurs between the time the process is planned and the time the process is actually performed; collecting at least one of the speech recognition results and text information of voices exchanged between the on-site personnel at the site where the process identified as an analysis target was performed and other people before and during the execution of the process identified as an analysis target; and generating a reason for the difference that occurred in the process identified as an analysis target based on at least one of the speech recognition results and text information, and outputting the generated reason.
[0010] One embodiment of the program causes a computer to perform the following steps: identify a process among several processes in airport ramp operation work that has a difference of a predetermined amount of time or more between the time the process is planned and the time the process is actually performed, as a target for analysis; collect at least one of the speech recognition results and text information of the voices exchanged between the on-site personnel at the site where the identified target for analysis process was performed and other people, before and during the execution of the identified target for analysis process; and generate a reason for the difference in the identified target for analysis process based on the speech recognition results and at least one of the text information, and output the generated reason.
[0011] According to the above-described embodiment, the effect is obtained that an information processing device, an information processing method, and a program can be provided that can improve the efficiency of lamp operation work.
[0012] This is a block diagram showing a schematic configuration example of the information processing device related to this disclosure. This is a flowchart illustrating an example of the schematic operation flow of the information processing device related to this disclosure. This is a block diagram showing a schematic configuration example of the information processing device related to this disclosure. This is a sequence diagram illustrating an example of the schematic operation flow of the information processing device related to this disclosure. This is a diagram illustrating a specific example of operation of the information processing device related to this disclosure. This is a diagram illustrating an example of a GUI screen output from the information processing device related to this disclosure. This is a block diagram showing a schematic configuration example of the information processing device related to this disclosure. This is a sequence diagram illustrating an example of the schematic operation flow of the information processing device related to this disclosure. This is a diagram illustrating a specific example of operation of the information processing device related to this disclosure. This is a diagram illustrating an example of a GUI screen output from the information processing device related to this disclosure. This is a block diagram showing a schematic hardware configuration example of the computer that implements the information processing device related to this disclosure.
[0013] Embodiments of this disclosure will be described below with reference to the drawings. Note that the following description and drawings have been omitted and simplified as appropriate for clarity of explanation. Furthermore, the same elements are denoted by the same reference numerals in the following drawings, and redundant explanations have been omitted where necessary. Also, the specific numerical values shown below are merely examples to facilitate understanding of this disclosure and are not limiting.
[0014] <Embodiment 1> First, the configuration of the information processing device 10 according to the present disclosure will be described. Figure 1 is a block diagram showing a schematic configuration example of the information processing device 10 according to the present disclosure. As shown in Figure 1, the information processing device 10 includes an analysis target identification unit 11, a collection unit 12, and a reason generation unit 13.
[0015] The analysis target identification unit 11 identifies, as the analysis target, the process in which a difference of a predetermined time or more occurs between the time the process was planned and the time the process was actually carried out, among the multiple processes in ramp operation operations at the airport. The multiple processes in ramp operation operations are the processes for taking off aircraft.
[0016] The collection unit 12 collects at least one of the following: speech recognition results of voices exchanged between on-site personnel at the site where the process identified as the target of analysis is performed and other people, and text information such as messages, before and during the execution of the process identified as the target of analysis.
[0017] The reason generation unit 13 generates the reason for the difference in the process identified as the target of analysis, based on at least one of the speech recognition result and text information, and outputs the generated reason.
[0018] Next, the operation of the information processing device 10 according to this disclosure will be described. Figure 2 is a flowchart illustrating an example of the general operation flow of the information processing device 10 according to this disclosure.
[0019] As shown in Figure 2, first, the analysis target identification unit 11 identifies, among the multiple processes in airport ramp operation work, the processes in which a difference of a predetermined time or more occurs between the time the process was planned and the time the process was actually carried out, as the analysis target (step S101).
[0020] Next, the collection unit 12 collects at least one of the speech recognition results and text information of the voices exchanged between the on-site personnel at the site where the process identified as the target of analysis was performed and other people, before and during the execution of the process identified as the target of analysis (step S102).
[0021] Subsequently, the reason generation unit 13 generates the reason for the difference in the process identified as the target of analysis based on the speech recognition result and at least one of the text information, and outputs the generated reason (step S103).
[0022] According to this embodiment 1, it is possible to obtain the reason for any difference in time from the planned value for any of the processes in the ramp operation work, thereby improving the efficiency of the ramp operation work.
[0023] The reason generation unit 13 may also include a reason generation model. The reason generation unit 13 may input at least one of the speech recognition result and text information into the reason generation model and obtain the reason for the difference that occurred in the process identified as the target of analysis as the output of the reason generation model.
[0024] Furthermore, the reason generation unit 13 may collect expressions such as delay and advancement, as well as similar expressions, from at least one of the speech recognition results and text information, and input a prompt to the reason generation model instructing it to analyze the reasons why a difference occurred in the process identified as the target of analysis based on the collected content.
[0025] Furthermore, the analysis target identification unit 11 may exclude from the analysis any process that is performed after the process identified as the analysis target and that is in a critical path relationship with the process identified as the analysis target.
[0026] The information processing device 10 may also further include an aggregation unit, a summarization target identification unit, and a summary generation unit. Here, the aggregation unit may aggregate the reasons why differences occurred in the processes identified as analysis targets for multiple aircraft. The summarization target identification unit may, based on the aggregation results, identify as summarization targets the processes among the analysis targets for which differences occurred at a predetermined frequency or higher. The summary generation unit may generate a summary of the reasons why differences occurred in the processes identified as summarization targets based on the aggregation results for the processes identified as summarization targets, and output the generated summary.
[0027] Furthermore, the summary generation unit may also include a summary generation model. The summary generation unit may input the aggregated results related to the process identified as the target of summarization from the aggregated results into the summary generation model, and obtain a summary of the reasons why differences occurred in the process identified as the target of summarization as output of the summary generation model.
[0028] Furthermore, the summarization generation unit may input a prompt to the summarization generation model instructing it to analyze a summary of the reasons for differences occurring in the processes identified as targets for summarization, based on the aggregated results for those processes.
[0029] Furthermore, the multiple aircraft may be any combination of aircraft with the same flight number that took off from the same parking area within the airport on different days, a combination of aircraft with different flight numbers that took off from the same parking area within the airport on the same day, or a combination of aircraft with the same or different flight numbers that took off from any of the multiple parking areas within the airport on the same day or on different days.
[0030] <Embodiment 2> Embodiment 2 and Embodiments 3 and 4, described later, are embodiments that further elaborate on Embodiment 1 described above.
[0031] Furthermore, in this embodiment 2 and embodiments 3 and 4 described later, the multiple processes in ramp operation work at the airport are assumed to be the following six processes.
[0032] (1) In-block / Off-block: In-block refers to the aircraft entering the parking area, and off-block refers to the aircraft leaving the parking area. In the in-block / off-block process, the start time is the time the aircraft enters the parking area, and the end time is the time the aircraft leaves the parking area.
[0033] (2) Choke installation / removal. Chokes are wheel chocks for aircraft. In the choke installation / removal process, the start time is the time the chokes are installed on the aircraft, and the end time is the time the chokes are removed from the aircraft.
[0034] (3) Bridge Connection A bridge is a facility that connects the airport terminal building to an aircraft and allows passengers and crew to board and disembark from the aircraft. In the bridge connection process, the start time is the time when the bridge is connected to the aircraft, and the end time is the time when the connection between the bridge and the aircraft is released.
[0035] (4) Refueling In the refueling process, the start time is the time when refueling of the aircraft begins, and the end time is the time when refueling is completed.
[0036] (5) Cargo Doors A cargo door is a door that leads to the cargo compartment where cargo is loaded inside the aircraft. In the cargo door process, the start time is the time when the cargo door is opened, and the end time is the time when the cargo door is closed after the cargo has been loaded into the cargo compartment.
[0037] (6) Completion of pushback car connection A pushback car is a vehicle used to move an aircraft backward. In the process of completing the pushback car connection, the start time is the time when the pushback car is connected to the aircraft, and the end time is the time when the connection between the pushback car and the aircraft is released after the aircraft has been moved backward.
[0038] First, the configuration of the information processing device 10X according to this disclosure will be described. Figure 3 is a block diagram showing a schematic configuration example of the information processing device 10X according to this disclosure. As shown in Figure 3, the information processing device 10X includes a plan storage unit 101, a performance analysis unit 102, an analysis target identification unit 103, a status collection unit 104, and a reason generation unit 105.
[0039] The planning and storage unit 101 stores the planned start time, end time, and work time for each of the six processes. The work time for a process is the time from the start time to the end time of that process. The planning and storage unit 101 then outputs the planned start time, end time, and work time for each of the six processes to the analysis target identification unit 103.
[0040] The performance analysis unit 102 analyzes the actual start time, end time, and working time for each of the six processes, based on video footage taken by a camera of the site where the process was carried out. The performance analysis unit 102 then outputs the actual start time, end time, and working time for each of the six processes to the analysis target identification unit 103.
[0041] The analysis target identification unit 103 calculates the difference for each of the six processes by subtracting the planned values for the start time, end time, and work time from the actual values for each of these values. If, among the six processes, there is a process where the difference for at least one of the start time, end time, and work time is positive and exceeds a predetermined time, the analysis target identification unit 103 identifies that process as the analysis target for which the reason for the difference will be generated. The analysis target identification unit 103 then outputs the process identified as the analysis target, along with the actual values for the start time, end time, and work time related to the analysis target, to the reason generation unit 105. The analysis target identification unit 103 also outputs the process identified as the analysis target to the status collection unit 104.
[0042] The situation collection unit 104 collects at least one of the following: the speech recognition result of the voice exchanged between the on-site personnel at the site where the process identified as the target of analysis was performed and other people, and text information such as messages. For example, the other people with whom the on-site personnel interact may be the airline operator of the aircraft taking off, or other on-site personnel. The speech recognition result is the recognition result of the voice exchanged by the on-site personnel using a communication device such as a smartphone, and the text information is messages exchanged by the on-site personnel using a communication terminal such as a tablet. The situation collection unit 104 then outputs at least one of the speech recognition result and text information related to the process identified as the target of analysis to the reason generation unit 105.
[0043] In the following description and in embodiments 3 and 4 described later, for the sake of simplicity, the situation collection unit 104 collects both speech recognition results and text information, and outputs the collected speech recognition results and text information to the reason generation unit 105.
[0044] The reason generation unit 105 includes a reason generation AI (Artificial Intelligence) 106. The reason generation AI 106 is a generation model that generates the reasons for the differences occurring in the process specified as the analysis target, and is realized by an LLM (Large Language Model) or the like.
[0045] The reason generation unit 105 extracts the voice recognition results and text information that were exchanged before and during the implementation of the process, from the voice recognition results and text information related to the process, based on the actual values of the start time and end time related to the process specified as the analysis target. Specifically, the reason generation unit 105 extracts the voice recognition results and text information that were exchanged from a certain time before the actual value of the start time to the actual value of the end time.
[0046] Then, the reason generation unit 105 inputs the process specified as the analysis target, and the voice recognition results and text information before and during the implementation of the process, to the reason generation AI 106.
[0047] Further, the reason generation unit 105 collects expressions such as "delay" and "advance" and similar expressions similar to these expressions from the voice recognition results and text information, and inputs a prompt to the reason generation AI 106 to instruct it to analyze the reasons for the differences occurring in the process specified as the analysis target from the collected content. For example, expressions similar to "delay" and "advance" may include expressions such as delay, extension, trouble, obstacle, speed up, early, finish early, etc.
[0048] Then, the reason generation AI 106 collects expressions such as "delay" and "advance" and similar expressions similar to these expressions from the voice recognition results and text information according to the prompt, and analyzes the reasons for the differences occurring in the process specified as the analysis target from the collected content, and outputs the analysis result. For example, if there is voice recognition result or text information such as "It takes time for the cargo inspection and the loading is delayed by about 10 minutes...", the reason generation AI 106 collects this and uses it for the analysis of the reason.
[0049] Therefore, the reason generation unit 105 obtains the reason for the difference that occurred in the process identified as the target of analysis as the output of the reason generation AI 106. The reason generation unit 105 then outputs the reason for the difference that occurred in the process identified as the target of analysis to the outside. For example, the reason generation unit 105 outputs to the outside a GUI (Graphical User Interface) screen that describes the reason for the difference that occurred in the process identified as the target of analysis.
[0050] In this example, the reason generation unit 105 extracts speech recognition results and text information exchanged before and during the execution of the process from among the speech recognition results and text information, and inputs the extracted speech recognition results and text information to the reason generation AI 106, but this is not limited to this. For example, the reason generation unit 105 may input the speech recognition results and text information as is to the reason generation AI 106, and also input actual values of the start time and end time to the reason generation AI 106. The reason generation AI 106 may then extract speech recognition results and text information exchanged before and during the execution of the process from among the speech recognition results and text information. In this case, the prompt to be input to the reason generation AI 106 may be, for example, a prompt that instructs the AI to extract speech recognition results and text information exchanged before and during the execution of the process identified as the target of analysis from the speech recognition results and text information, collect expressions such as "delay" and "bring forward" and similar expressions from the extracted speech recognition results and text information, and analyze the reasons why a difference occurred in that process based on the collected content.
[0051] Next, the operation of the information processing device 10X according to this disclosure will be described. First, the operation flow of the information processing device 10X according to this disclosure will be described. Figure 4 is a sequence diagram illustrating an example of the general operation flow of the information processing device 10X according to this disclosure. In Figure 4, it is assumed that the plan storage unit 101 stores the planned start time, end time, and work time for each of the six processes.
[0052] As shown in Figure 4, first, the planning storage unit 101 outputs the planned start time, end time, and work time for each of the six processes to the analysis target identification unit 103 (step S201). Then, the actual analysis unit 102 analyzes the actual start time, end time, and work time for each of the six processes (step S202), and outputs the analyzed actual values to the analysis target identification unit 103 (step S203).
[0053] Next, the analysis target identification unit 103 calculates the difference for each of the six processes by subtracting the planned values for the start time, end time, and work time from the actual values for each of these values. Next, the analysis target identification unit 103 identifies the processes among the six processes in which the difference between at least one of the start time, end time, and work time is positive and is equal to or greater than a predetermined time as the analysis target (step S204). Next, the analysis target identification unit 103 outputs the processes identified as analysis targets to the status collection unit 104 (step S205). The analysis target identification unit 103 also outputs the processes identified as analysis targets, along with the actual values for the start time, end time, and work time related to those processes, to the reason generation unit 105 (step S206).
[0054] Next, the situation collection unit 104 collects speech recognition results and text information of conversations between on-site personnel at the site where the identified process was carried out and other people (step S207), and outputs the collected speech recognition results and text information to the reason generation unit 105 (step S208).
[0055] Next, the reason generation unit 105 extracts speech recognition results and text information exchanged before and during the execution of the process identified as the target of analysis. Next, the reason generation unit 105 inputs a prompt to the reason generation AI 106, along with the process identified as the target of analysis, and the speech recognition results and text information before and during the execution of that process. As a result, the reason generation unit 105 obtains the reason for the difference in the process identified as the target of analysis as the output of the reason generation AI 106 (step S209). Subsequently, the reason generation unit 105 outputs the reason for the difference in the process identified as the target of analysis to the outside (step S210).
[0056] Next, the specific operation of the information processing device 10X related to this disclosure will be explained. Figure 5 is a diagram illustrating a specific example of the operation of the information processing device 10X related to this disclosure. In Figure 5, it is assumed that the predetermined time for the difference obtained by subtracting the planned value from the actual value is 10 minutes.
[0057] In the example in Figure 5, the difference in completion time for the choke installation / removal process is +10 minutes. Also, in the cargo door process, the difference in completion time is +20 minutes, and the difference in working time is +18 minutes. Furthermore, in the pushback car connection completion process, the difference in start time is +10 minutes.
[0058] Therefore, the analysis target identification unit 103 identifies three of the six processes as the analysis targets: the choke installation / removal process, the cargo door process, and the pushback car connection completion process. Accordingly, the reason generation unit 105 then generates the reasons for the differences that occurred for each of the three processes identified as the analysis targets, and outputs the generated reasons.
[0059] Next, we will describe the GUI screen output from the information processing device 10X related to this disclosure. Figure 6 is a diagram illustrating an example of the GUI screen output from the information processing device 10X related to this disclosure. Note that Figure 6 is an example of the GUI screen resulting from Figure 5.
[0060] In the example shown in Figure 5, three processes are identified as the targets of analysis: the choke installation / removal process, the cargo door process, and the pushback car connection completion process. Therefore, the GUI screen shown in Figure 6 describes the reason for the difference for each of the three processes identified as the targets of analysis.
[0061] As described above, according to this embodiment 2, the analysis target identification unit 103 calculates the difference for each of the six processes by subtracting the planned values for the start time, end time, and work time from the actual values for each of the start time, end time, and work time, and identifies processes where the difference for at least one of the start time, end time, and work time is greater than or equal to a predetermined time as the analysis target. The situation collection unit 104 collects speech recognition results and text information of voices exchanged between the on-site personnel at the site where the identified analysis target processes were performed and other people. The reason generation unit 105 inputs the speech recognition results and text information exchanged before and during the implementation of the identified analysis target processes to the reason generation AI 106, obtains the reason for the difference that occurred in the identified analysis target processes, and outputs that reason. As a result, it is possible to obtain the reason for the difference that occurred in processes among the multiple processes in the ramp operation work where the difference from the planned value is greater than a predetermined time, thereby improving the efficiency of the ramp operation work.
[0062] <Embodiment 3> First, the configuration of the information processing device 10Y according to the present disclosure will be described. Figure 7 is a block diagram showing a schematic configuration example of the information processing device 10Y according to the present disclosure. As shown in Figure 7, the information processing device 10Y differs from the information processing device 10X described above in that the planning storage unit 101 and the analysis target identification unit 103 are replaced by the planning storage unit 101A and the analysis target identification unit 103A.
[0063] The planning and storage unit 101A stores the planned start time, end time, and work time for each of the six processes, as well as the critical path relationships between the six processes. For example, if a subsequent process cannot start or end until the previous process is completed, then the previous and subsequent processes are in a critical path relationship. The planning and storage unit 101A then outputs the planned start time, end time, and work time for each of the six processes, along with the critical path relationships between the six processes, to the analysis target identification unit 103A.
[0064] The analysis target identification unit 103A calculates the difference for each of the six processes by subtracting the planned values for the start time, end time, and work time from the actual values for each of these values. The analysis target identification unit 103A then extracts from the six processes the processes in which the difference of at least one of the start time, end time, and work time is positive and is equal to or greater than a predetermined time. The analysis target identification unit 103A then identifies the processes to be analyzed from the extracted processes based on the critical path relationship. Specifically, first, the analysis target identification unit 103A identifies the first process among the extracted processes as the analysis target. Subsequently, the analysis target identification unit 103A excludes processes from the analysis target that are performed after the process identified as the analysis target and that are in a critical path relationship with the process identified as the analysis target, while identifying the other processes as analysis targets. The analysis target identification unit 103A then outputs the process identified as the analysis target, along with the start time, end time, and actual working time for that process, to the reason generation unit 105. The analysis target identification unit 103A also outputs the process identified as the analysis target to the status collection unit 104. The other configurations of the information processing device 10Y are the same as those of the information processing device 10X described above.
[0065] Next, the operation of the information processing device 10Y according to this disclosure will be described. First, the operation flow of the information processing device 10Y according to this disclosure will be described. Figure 8 is a sequence diagram illustrating an example of the general operation flow of the information processing device 10Y according to this disclosure. In Figure 8, it is assumed that the planning storage unit 101A stores the planned start time, end time, and work time for each of the six processes, as well as the critical path relationship between the six processes.
[0066] As shown in Figure 8, first, the planning and storage unit 101A outputs the planned start time, end time, and work time for each of the six processes, as well as the critical path relationship between the six processes, to the analysis target identification unit 103 (step S301). Then, the same processing as steps S202 and S203 in Figure 4 is performed in steps S302 and S303.
[0067] Next, the analysis target identification unit 103A calculates the difference for each of the six processes by subtracting the planned values for the start time, end time, and work time from the actual values for each of these values. Next, the analysis target identification unit 103A extracts the processes from the six processes in which the difference of at least one of the start time, end time, and work time is positive and is equal to or greater than a predetermined time. Next, the analysis target identification unit 103A identifies the processes to be analyzed from the extracted processes based on the critical path relationship (step S304). At this time, the analysis target identification unit 103A excludes processes from the analysis target that are performed after the processes identified as the analysis target and that are in a critical path relationship with the processes identified as the analysis target. Subsequently, the same processing as steps S205 to S210 in Figure 4 is performed in steps S305 to S310.
[0068] Next, the specific operation of the information processing device 10Y related to this disclosure will be explained. Figure 9 is a diagram illustrating a specific example of the operation of the information processing device 10Y related to this disclosure. In Figure 9, the predetermined time for the difference obtained by subtracting the planned value from the actual value is assumed to be 10 minutes.
[0069] In the example in Figure 9, similar to Figure 5, the difference in completion time for the choke installation / removal process is +10 minutes. Also, in the cargo door process, the difference in completion time is +20 minutes, and the difference in working time is +18 minutes. Furthermore, in the process of completing the connection of the pushback car, the difference in start time is +10 minutes.
[0070] Therefore, the analysis target identification unit 103A extracts three of the six processes: the choke installation / removal process, the cargo door process, and the pushback car connection completion process.
[0071] Furthermore, in the example shown in Figure 9, the three extracted processes are in a critical path relationship, with the cargo door process being performed first (1st), the choke installation / removal process being performed second, and the pushback car connection completion process being performed third.
[0072] Therefore, the analysis target identification unit 103A identifies the cargo door process, which is the first of the three extracted processes, as the target of analysis. However, the choke installation / removal process, which is the second process, and the pushback car connection completion process, which is the third process, are in a critical path relationship with the cargo door process. Therefore, the analysis target identification unit 103A excludes the choke installation / removal process and the pushback car connection completion process from the analysis target.
[0073] Next, we will describe the GUI screen output from the information processing device 10Y related to this disclosure. Figure 10 is a diagram illustrating an example of a GUI screen output from the information processing device 10Y related to this disclosure. Note that Figure 10 is an example of a GUI screen resulting from the results in Figure 9.
[0074] In the example shown in Figure 9, the cargo door process is identified as the target of analysis. Therefore, the GUI screen shown in Figure 10 describes the reason for the difference in the cargo door process, which has been identified as the target of analysis.
[0075] As described above, according to this embodiment 3, the analysis target identification unit 103A excludes from the analysis process any process that is performed after the process identified as the analysis target and is in a critical path relationship with the process identified as the analysis target. As a result, the number of processes to be analyzed, i.e., the number of processes for which the reason for the difference occurred, is reduced, thereby reducing the processing load on the information processing device 10Y. The other effects of this embodiment 3 are the same as those of the embodiment 2 described above.
[0076] <Embodiment 4> First, the configuration of the information processing device 10Z according to the present disclosure will be described. Figure 11 is a block diagram showing a schematic configuration example of the information processing device 10Z according to the present disclosure. As shown in Figure 11, the information processing device 10Z differs from the information processing device 10X described above in that a summarization unit 107, a summarization target identification unit 108, and a summary generation unit 109 are added.
[0077] In embodiments 2 and 3 described above, the process to be analyzed was identified for one aircraft, and the reasons for any differences in that process were generated. In contrast, in embodiment 4, the process to be analyzed is identified for each of multiple aircraft, and the reasons for any differences in that process are generated.
[0078] Here, the multiple aircraft may be any combination of aircraft with the same flight number that took off from the same parking area within the airport on different days, a combination of aircraft with different flight numbers that took off from the same parking area within the airport on the same day, or a combination of aircraft with the same or different flight numbers that took off from any of the multiple parking areas within the airport on the same day or on different days.
[0079] The aggregation unit 107 aggregates the reasons for differences in the processes identified as the target of analysis for each of the multiple aircraft. The aggregation unit 107 then outputs the aggregation results to the summarization target identification unit 108.
[0080] The summarization target identification unit 108 determines the frequency of occurrence of differences for each of the processes identified as analysis targets based on the aggregation results. Here, the frequency of occurrence is the frequency at which the process is identified as an analysis target. If there is a process among the processes identified as analysis targets that has a difference occurring at a frequency of a predetermined number of times or more, the summarization target identification unit 108 identifies that process as a summarization target for which a summary of the reasons for the difference will be generated. The summarization target identification unit 108 then outputs the process identified as a summarization target and the aggregation results related to that process to the summary generation unit 109. Here, the aggregation results are the aggregation results of the reasons for the differences in that process.
[0081] The summary generation unit 109 includes a summary generation AI 110. The summary generation AI 110 is a generation model that generates a summary of the reasons why differences occurred in the processes identified as the target of summarization, and is implemented by LLM or the like.
[0082] The summary generation unit 109 inputs the process identified as the target of summarization and the aggregated results related to that process to the summary generation AI 110. Furthermore, the summary generation unit 109 inputs a prompt to the summary generation AI 110 instructing it to analyze a summary of the reasons why differences occurred in the process identified as the target of summarization, based on the aggregated results related to that process.
[0083] Then, following the prompt, the summary generation AI 110 analyzes the aggregated results for the process identified as the target of the summary, summarizing the reasons why differences occurred in that process, and outputs the analysis results.
[0084] Therefore, the summary generation unit 109 obtains a summary of the reasons why differences occurred in the process identified as the target of summarization as output of the summary generation AI 110. The summary generation unit 109 then outputs the summary of the reasons why differences occurred in the process identified as the target of summarization to the outside. For example, the summary generation unit 109 outputs a GUI screen describing the reasons why differences occurred in the process identified as the target of summarization to the outside. The other configurations of the information processing device 10Z are the same as those of the information processing device 10X described above.
[0085] Next, the operation of the information processing device 10Z according to this disclosure will be described. First, the operation flow of the information processing device 10Z according to this disclosure will be described. Figure 12 is a sequence diagram illustrating an example of the general operation flow of the information processing device 10Z according to this disclosure. In Figure 12, it is assumed that the plan storage unit 101 stores the planned start time, end time, and work time for each of the six processes.
[0086] As shown in Figure 12, first, for each of the multiple aircraft, the same process as steps S201 to S210 in Figure 4, steps S401 to S410, is performed. This allows the reason for the difference in the process identified as the target of analysis to be obtained for each of the multiple aircraft. At this time, in step S410, the reason generation unit 105 outputs the reason for the difference to the aggregation unit 107, rather than outputting it externally.
[0087] Next, the aggregation unit 107 aggregates the reasons for differences in the processes identified as the target of analysis for each of the multiple aircraft (step S411). Then, the aggregation unit 107 outputs the aggregation results to the summarization target identification unit 108 (step S412).
[0088] Next, the summarization target identification unit 108 determines the frequency of occurrence of differences for each of the processes identified as analysis targets based on the aggregation results. Next, the summarization target identification unit 108 identifies the processes among the analysis targets for which differences have occurred at a predetermined frequency or higher as summarization targets (step S413). Next, the summarization target identification unit 108 outputs the processes identified as summarization targets and the aggregation results related to those processes to the summarization generation unit 109 (step S414).
[0089] Next, the summary generation unit 109 inputs a prompt to the summary generation AI 110, along with the process identified as the target of the summary, and the aggregated results related to that process. As a result, the summary generation unit 109 obtains a summary of the reasons why differences occurred in the process identified as the target of the summary as output of the summary generation AI 110 (step S415). Subsequently, the summary generation unit 109 outputs the summary of the reasons why differences occurred in the process identified as the target of the summary to the outside (step S416).
[0090] Next, we will describe the GUI screen output from the information processing device 10Z related to this disclosure. Figures 13 to 15 are diagrams illustrating examples of GUI screens output from the information processing device 10X related to this disclosure.
[0091] More specifically, Figure 13 illustrates an example of the GUI screen when analyzing an aircraft with the same flight number (N203) that took off on a different day from a single parking area (parking area 01). Figure 14 illustrates an example of the GUI screen when analyzing an aircraft with a different flight number that took off on the same day from the same parking area (parking area 01). Figure 15 illustrates an example of the GUI screen when analyzing an aircraft with a different flight number that took off on the same day from one of several parking areas (parking areas 01, 02, 03).
[0092] In the examples shown in Figures 13 to 15, the cargo door process is identified as the target of the summary. Therefore, in the GUI screens shown in Figures 13 to 15, the overall status column contains a summary of the reason why a difference occurred in the cargo door process.
[0093] Furthermore, the GUI screen shown in Figure 13 also describes the reasons for any differences in the cargo door process on a daily basis. The GUI screen shown in Figure 14 also describes the reasons for any differences in the cargo door process for each flight number. The GUI screen shown in Figure 15 also describes the reasons for any differences in the cargo door process for each parking location.
[0094] As described above, according to this embodiment 4, the aggregation unit 107 aggregates the reasons for differences occurring in the processes identified as analysis targets for each of the multiple aircraft. The summarization target identification unit 108 identifies processes among the analysis targets that have occurred at a predetermined frequency or more as summarization targets. The summary generation unit 109 inputs the aggregation results regarding the processes identified as summarization targets to the summary generation AI 110, obtains a summary of the reasons for differences occurring in the processes identified as summarization targets, and outputs the summary. This makes it possible to obtain a summary of the reasons for differences for processes among the multiple processes in ramp operation operations that frequently have a difference of a predetermined time or more from the planned value, thereby improving the efficiency of ramp operation operations. The other effects of this embodiment 4 are the same as those of embodiment 2 described above.
[0095] <Other Embodiments> In embodiments 1 to 4 described above, multiple components were provided within the information processing devices 10, 10X to 10Z, but the invention is not limited thereto. In this disclosure, these multiple components may be distributed across multiple devices. That is, this disclosure may be implemented by a system comprising multiple devices.
[0096] Furthermore, in embodiments 2 to 4 described above, the GUI screen output from the information processing devices 10X to 10Z contained a description of the reason for the difference or a summary of the reason for the difference, but this is not limited to this. In this disclosure, the GUI screen may also include video footage taken by a camera of the site where multiple processes in the lamp operation work were carried out, a list of information as shown in Figure 5 or Figure 9, a graph of the information as shown in Figure 5 or Figure 9, etc. Note that the graph may be a graph showing the difference that occurred in each of the multiple processes, but this is not limited to this.
[0097] Furthermore, in embodiments 2 to 4 described above, the information processing devices 10X to 10Z input speech recognition results, etc., related to a process in which a difference of a predetermined time or more occurred from the planned value into the reason generation AI 106 to obtain the reason for the difference in that process, but are not limited to this. In this disclosure, speech recognition results, etc., related to a process in which no difference of a predetermined time or more occurred from the planned value may be input into the reason generation AI 106 to obtain the reason for no difference in that process. Alternatively, in this disclosure, speech recognition results, etc., related to a process in which a difference of a predetermined time or more occurred from the planned value and speech recognition results, etc., related to a process in which no difference of a predetermined time or more occurred from the planned value may be input into the reason generation AI 106 to obtain the difference between the speech recognition results, etc., of both processes.
[0098] <Hardware Configuration of Information Processing Device> Figure 16 is a block diagram showing a schematic example of the hardware configuration of a computer 90 that implements the information processing devices 10, 10X to 10Z according to this disclosure.
[0099] As shown in Figure 16, the computer 90 includes a processor 91, memory 92, storage 93, input / output interface (input / output I / F) 94, and communication interface (communication I / F) 95. The processor 91, memory 92, storage 93, input / output interface 94, and communication interface 95 are connected to each other by a data transmission path for sending and receiving data.
[0100] The processor 91 is a processing unit such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit). The memory 92 is a memory such as RAM (Random Access Memory) or ROM (Read Only Memory). The storage 93 is a storage device such as an HDD (Hard Disk Drive), SSD (Solid State Drive), or memory card. The storage 93 may also be a memory such as RAM or ROM.
[0101] A program is stored in the storage 93. This program includes a set of instructions (or software code) that, when loaded into a computer, causes the computer 90 to perform one or more functions of the information processing devices 10, 10X to 10Z described above. The components of the information processing devices 10, 10X to 10Z described above may also be realized by the processor 91 loading and executing the program stored in the storage 93. Furthermore, the storage function of the information processing devices 10, 10X to 10Z described above may be realized by memory 92 or storage 93.
[0102] Furthermore, the programs described above may be stored on non-temporary computer-readable media or tangible storage media. Examples, but not limited to, include RAM, ROM, flash memory, SSD, or other memory technologies; CD (Compact Disc)-ROM, DVD (Digital Versatile Disc), Blu-ray® disc, or other optical disc storage; magnetic cassettes, magnetic tapes, magnetic disk storage, or other magnetic storage devices. The programs may also be transmitted over temporary computer-readable media or communication media. Examples, but not limited to, include temporary computer-readable media or communication media, including electrically, optically, acoustically, or otherwise propagating signals.
[0103] The input / output interface 94 is connected to a display device 941, an input device 942, a sound output device 943, and the like. The display device 941 is a device that displays a screen corresponding to the drawing data processed by the processor 91, such as an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube) display, or a monitor. The input device 942 is a device that receives operator input, such as a keyboard, mouse, and touch sensor. The display device 941 and the input device 942 may be integrated and implemented as a touch panel. The sound output device 943 is a device that outputs sound corresponding to the acoustic data processed by the processor 91, such as a speaker.
[0104] The communication interface 95 transmits and receives data to and from an external device. For example, the communication interface 95 communicates with an external device via a wired communication path or a wireless communication path.
[0105] Although the present disclosure has been described above with reference to embodiments, the present disclosure is not limited to the embodiments described above. Various modifications to the structure and details of the present disclosure can be made as can be understood by those skilled in the art within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.
[0106] Furthermore, each drawing is merely illustrative to illustrate one or more embodiments. Each drawing may be associated not only with one specific embodiment but also with one or more other embodiments. As those skilled in the art will understand, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings, for example, to create embodiments not explicitly shown or described. Not all features or steps shown in any one drawing to illustrate an exemplary embodiment are necessarily required, and some features or steps may be omitted. The order of steps shown in any of the drawings may be changed as appropriate.
[0107] Furthermore, some or all of the above embodiments may also be described as follows, but are not limited to the following: (Note 1) An information processing device comprising: an analysis target identification means for identifying an analysis target among a plurality of processes in airport ramp operation work in which a difference of a predetermined time or more occurs between the time the process is planned and the time the process is actually performed; a collection means for collecting at least one of the speech recognition results and text information of voices exchanged between the on-site personnel at the site where the analysis target was performed and other people before and during the performance of the analysis target identified process; and a reason generation means for generating the reason why the difference occurred in the analysis target identified process based on at least one of the speech recognition results and the text information, and outputting the generated reason. (Note 2) The information processing device according to Note 1, wherein the reason generation means comprises a reason generation model, inputs at least one of the speech recognition results and the text information into the reason generation model, and obtains the reason why the difference occurred in the analysis target identified process as the output of the reason generation model. (Note 3) The information processing apparatus according to Note 2, wherein the reason generation means collects expressions such as delay and advancement and similar expressions similar to those expressions from at least one of the speech recognition result and the text information, and inputs a prompt to the reason generation model instructing it to analyze the reason why the difference occurred in the process identified as the target of analysis based on the collected content. (Note 4) The information processing apparatus according to Note 1, wherein the analysis target identification means excludes from the analysis a process that is performed after the process identified as the target of analysis and is in a critical path relationship with the process identified as the target of analysis.(Note 5) The information processing device according to Note 1, comprising: an aggregation means for aggregating the reasons why the difference occurred in the process identified as the subject of analysis for each of a plurality of aircraft; a summarization target identification means for identifying, based on the aggregation results, processes among the processes identified as the subject of analysis in which the difference occurred at a predetermined frequency or more; and a summary generation means for generating a summary of the reasons why the difference occurred in the process identified as the subject of summary, based on the aggregation results relating to the process identified as the subject of summary, and outputting the generated summary. (Note 6) The information processing device according to Note 5, wherein the summary generation means comprises a summary generation model, inputs the aggregation results relating to the process identified as the subject of summary from the aggregation results into the summary generation model, and obtains a summary of the reasons why the difference occurred in the process identified as the subject of summary as output of the summary generation model. (Note 7) The information processing device according to Note 6, wherein the summary generation means inputs a prompt to the summary generation model instructing it to analyze a summary of the reasons why the difference occurred in the process identified as the target of the summary, from the aggregated results relating to the process identified as the target of the summary among the aggregated results. (Note 8) The information processing device according to Note 5, wherein the multiple aircraft are any of the following: a combination of the aircraft with the same flight number that took off on different days from the same parking area in the airport; a combination of the aircraft with different flight numbers that took off on the same day from the same parking area in the airport; or a combination of the aircraft with the same or different flight numbers that took off on the same day or on different days from any of the multiple parking areas in the airport.(Note 9) An information processing method performed by an information processing device, comprising: identifying, as a target for analysis, a process among a plurality of processes in airport ramp operation work, in which a difference of a predetermined time or more occurs between the time the process was planned and the time the process was actually performed; collecting at least one of the speech recognition results and text information of voices exchanged between the on-site personnel at the site where the process identified as the target for analysis was performed and other people before and during the execution of the process identified as the target for analysis; and generating a reason for the difference that occurred in the process identified as the target for analysis based on at least one of the speech recognition results and text information, and outputting the generated reason. (Note 10) A program that causes a computer to execute the following steps: a procedure for identifying, among multiple steps in airport ramp operation work, a step in which a difference of a predetermined amount of time or more occurs between the time the step is planned and the time the step is actually performed, as a target for analysis; a procedure for collecting at least one of the speech recognition results and text information of the voices exchanged between the on-site personnel at the site where the identified target for analysis was performed and other people, before and during the performance of the identified target for analysis; and a procedure for generating the reason why the difference occurred in the identified target for analysis based on the speech recognition results and at least one of the text information, and outputting the generated reason.
[0108] Furthermore, some or all of the elements (e.g., configuration and function) described in Appendices 2 to 8 that are subordinate to Appendice 1 may also be subordinate to Appendices 9 and 10 in the same manner as those described in Appendices 2 to 8. Some or all of the elements described in any appendice may be applied to various hardware, software, recording means, systems, and methods for recording software.
[0109] This application claims priority based on Japanese Patent Application No. 2025-018515, filed on 6 February 2025, and incorporates all of its disclosures herein.
[0110] 10 Information Processing Device 11 Analysis Target Identification Unit 12 Collection Unit 13 Reason Generation Unit 10X-10Z Information Processing Device 101, 101A Plan Storage Unit 102 Performance Analysis Unit 103, 103A Analysis Target Identification Unit 104 Situation Collection Unit 105 Reason Generation Unit 106 Reason Generation AI 107 Aggregation Unit 108 Summarization Target Identification Unit 109 Summarization Generation Unit 110 Summarization Generation AI 90 Computer 91 Processor 92 Memory 93 Storage 94 Input / Output Interface 941 Display Device 942 Input Device 943 Sound Output Device 95 Communication Interface
Claims
1. An information processing device comprising: an analysis target identification means for identifying, among multiple processes in airport ramp operation work, a process in which a difference of a predetermined time or more occurs between the time the process is planned and the time the process is actually performed; a collection means for collecting at least one of the speech recognition results and text information of voices exchanged between the on-site personnel at the site where the process identified as the analysis target was performed and other people, before and during the performance of the process identified as the analysis target; and a reason generation means for generating the reason why the difference occurred in the process identified as the analysis target based on at least one of the speech recognition results and text information, and outputting the generated reason.
2. The information processing apparatus according to claim 1, wherein the reason generation means comprises a reason generation model, inputs at least one of the speech recognition result and the text information into the reason generation model, and obtains as the output of the reason generation model the reason why the difference occurred in the process identified as the target of analysis.
3. The information processing apparatus according to claim 2, wherein the reason generation means collects expressions such as delay and advancement, and similar expressions similar to those expressions, from at least one of the speech recognition result and the text information, and inputs a prompt to the reason generation model instructing it to analyze the reason why the difference occurred in the process identified as the target of analysis, based on the collected content.
4. The information processing apparatus according to claim 1, wherein the means for identifying the object of analysis excludes from the object of analysis any process that is performed after the process identified as the object of analysis and is in a critical path relationship with the process identified as the object of analysis.
5. Information processing apparatus according to claim 1, comprising: an aggregation means for aggregating the reasons for the occurrence of the difference in the process identified as the subject of analysis for each of a plurality of aircraft; a summarization target identification means for identifying, based on the aggregation results, processes among the processes identified as the subject of analysis in which the occurrence of the difference has occurred at a predetermined frequency or higher, as the subject of summarization; and a summary generation means for generating a summary of the reasons for the occurrence of the difference in the process identified as the subject of summarization, based on the aggregation results relating to the process identified as the subject of summarization, and outputting the generated summary.
6. The information processing apparatus according to claim 5, wherein the summary generation means comprises a summary generation model, inputs the aggregated results relating to the process identified as the target of the summary from the aggregated results into the summary generation model, and obtains a summary of the reasons why the difference occurred in the process identified as the target of the summary as output of the summary generation model.
7. The information processing apparatus according to claim 6, wherein the summary generation means inputs a prompt to the summary generation model instructing it to analyze a summary of the reasons why the difference occurred in the process identified as the target of the summary, based on the aggregated results relating to the process identified as the target of the summary from the aggregated results relating to the process identified as the target of the summary.
8. The information processing device according to claim 5, wherein the plurality of aircraft are any combination of aircraft with the same flight number that took off on different days from the same parking area within the airport, a combination of aircraft with different flight numbers that took off on the same day from the same parking area within the airport, or a combination of aircraft with the same or different flight numbers that took off on the same day or on different days from any of the plurality of parking areas within the airport.
9. An information processing method performed by an information processing device, comprising: identifying, as a target for analysis, a process among a plurality of processes in airport ramp operation work, in which a difference of a predetermined time or more occurs between the time the process was planned and the time the process was actually performed; collecting at least one of the speech recognition results and text information of voices exchanged between the on-site personnel at the site where the process identified as the target for analysis was performed and other people before and during the execution of the process identified as the target for analysis; and generating a reason for the difference that occurred in the process identified as the target for analysis based on at least one of the speech recognition results and text information, and outputting the generated reason.
10. A program that causes a computer to execute the following steps: a procedure for identifying, among multiple steps in airport ramp operation work, a step in which a difference of a predetermined amount of time or more occurs between the time the step is planned and the time the step is actually performed, as a subject for analysis; a procedure for collecting at least one of the speech recognition results and text information of conversations between the on-site personnel at the site where the identified subject for analysis was performed and other people, before and during the execution of the identified subject for analysis; and a procedure for generating the reason why the difference occurred in the identified subject for analysis based on at least one of the speech recognition results and text information, and outputting the generated reason.