Operation management assistance system, operation management assistance method, and computer-readable storage medium

Abstract

The operation management support system can access a first set of associated data that links biometric data related to the driver's body and business status data related to the driver's business status, and a second set of hazard assessment results representing the hazard of the driver's driving. The operation management support system performs: a first acquisition process, acquiring a group of associated data related to the driver's specific business status from the first set according to business status, and acquiring a specific hazard assessment result group for the driver's specific business status from the second set; and a generation process, using the associated data group related to the specific business status acquired through the first acquisition process and the specific hazard assessment result group for the specific business status, generating a presumption model according to the specific business status, and storing it in a third set, the presumption model presumpturing the driver's accident risk in the specific business status.

Description

[0001] Reference-based introduction

[0002] This application claims priority to Japanese application filed on June 8, 2021, namely Japanese Patent Application No. 2021-095970, the contents of which are incorporated herein by reference. Technical Field

[0003] This invention relates to an operation management auxiliary system and an operation management auxiliary method. Background Technology

[0004] In recent years, traffic accidents caused by driver fatigue in long-distance commercial vehicles such as logistics trucks and night buses have become a significant social problem. To prevent such accidents, technologies using biosensors to monitor the health of drivers during driving are constantly evolving.

[0005] In particular, the development of technologies for estimating autonomic neural function (ANF) by applying time-frequency domain analysis based on heart rate intervals over a short time span, analyzing the estimated ANF to infer the driver's health status, fatigue level, and the risk of accidents primarily caused by these factors, and providing feedback of the estimated results to the driver or their manager is ongoing. Currently, many technologies have been developed for measuring or analyzing biological data under specific, continuous conditions, such as during rest or driving.

[0006] For example, Patent Document 1 discloses a car customer service system that detects the user's state and, based on the detection results, autonomously controls the operation of the in-vehicle devices in the manner most desired by the user. Regarding this car customer service system, the action content of the reception action unit varies according to the content of the user's biometric information, thereby further appropriatening the service (reception) effect when using the car based on the user's mental or physical state. Specifically, by extracting reference information for action control of functions determined according to a function extraction matrix, and adding the physical or mental state reflected by additionally acquired user biometric information to this reference information, appropriateness of the action content of the selected functions can be achieved.

[0007] Existing technical documents

[0008] Patent documents

[0009] Patent Document 1: Japanese Patent Application Publication No. 2008-126818 Summary of the Invention

[0010] The problem that the invention aims to solve

[0011] The technology of Patent Document 1 is applied to scenarios where accident risks are estimated and feedback is provided. For example, consider the following situation: the scenario is estimated as "driving," the user's physical / mental state is estimated as "drowsy," and the purpose of driving includes "preventing drowsiness (improving safety)." In this case, the risk of "drowsy driving" in the considered accident can be reduced.

[0012] However, the main causes of accidents involving drivers of commercial vehicles, even when limited to the driver's biological state, include not only drowsiness but also various other biological states such as decreased concentration, excitement, and excessive tension. Furthermore, the types of accidents also involve "distracted driving," "reckless driving," and many other scenarios where accidents or dangerous situations are caused by multiple biological factors. Therefore, it is difficult to clearly define the causal relationship between each biological factor and the accident. Thus, even with detailed descriptions of the purpose and presumed physical / mental state in the technology of Patent Document 1, it is difficult to achieve the estimation and feedback of accident risk.

[0013] The purpose of this invention is to achieve accident risk estimation corresponding to multiple different scenarios.

[0014] Methods for solving problems

[0015] An operation management assistance system, as one aspect of the invention disclosed in this application, comprises: a processor that executes a program; and a storage device that stores the program, wherein the processor is capable of accessing a first set of association data that links biometric data related to the driver's body and business status data related to the driver's business status, and a second set of hazard assessment results representing the hazard of the driver's driving; the processor performs the following processes: a first acquisition process that acquires a set of association data related to a specific business status of the driver from the first set, categorized by business status, and acquires a set of specific hazard assessment results for the specific business status of the driver from the second set; and a generation process that uses the set of association data related to the specific business status and the set of specific hazard assessment results obtained through the first acquisition process to generate a presumption model categorized by the specific business status and stores it in a third set, the presumption model presumpturing the accident risk of the driver in the specific business status.

[0016] Invention Effects

[0017] According to representative embodiments of the present invention, it is possible to estimate accident risks corresponding to multiple different scenarios. Other issues, structures, and effects beyond those described above will become clear through the following description of embodiments. Attached Figure Description

[0018] Figure 1 This is an explanatory diagram illustrating an example of the system structure of an operation management support system.

[0019] Figure 2 This is a block diagram representing an example of the hardware structure of an operation management auxiliary system.

[0020] Figure 3 It is a flowchart of a series of processes in the operation and management support system, from the acquisition of biological measurement data to the estimation of accident risk and the presentation of results.

[0021] Figure 4 This is an explanatory diagram illustrating a structural example of biological measurement data.

[0022] Figure 5 This is an explanatory diagram illustrating a structural example of business status data.

[0023] Figure 6 This is an explanatory diagram illustrating a structural example of a danger assessment database (DB).

[0024] Figure 7 This is a flowchart illustrating an example of the process for acquiring and processing biological measurement-business status DB or biological measurement-business status data.

[0025] Figure 8 This is a chart illustrating examples of heart rate fluctuations.

[0026] Figure 9 It is a graph showing the time variation of RR.

[0027] Figure 10 This is a graph showing the spectral variation of RR over time.

[0028] Figure 11 This is an explanatory diagram illustrating an example of processing biological measurement-business status data obtained from biological measurement data and business status data.

[0029] Figure 12 This is an explanatory diagram illustrating a structural example of acquired biological measurement-operational status data.

[0030] Figure 13 It means Figure 3 The flowchart shows a detailed example of the process for generating or updating the model of estimated accident risk (step S103).

[0031] Figure 14 This is an explanatory diagram showing the training examples of an inference model that uses data segmented by business status labels.

[0032] Figure 15 It means Figure 3The flowchart shown is a detailed example of the accident risk estimation process (step S104) based on the accident risk estimation procedure.

[0033] Figure 16 It means Figure 3 The estimated result shown is a flowchart of a detailed processing example of the process (step S105).

[0034] Figure 17 This is an explanatory diagram illustrating an example of accident risk estimation processing (step S104).

[0035] Figure 18 This is an explanatory diagram illustrating an example of feedback provided by prompting the processing of the presumed result (step S105) when the presumed result is "dangerous" or based on such a situation.

[0036] Figure 19 This is an illustrative diagram illustrating an example of feedback where, after the day's business has concluded and biometric data has been obtained, or the following day, drivers and managers can refer to the shift in past accident risks. Detailed Implementation

[0037] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0038] Example 1

[0039] <System Structure Example>

[0040] Figure 1 This is an explanatory diagram illustrating an example of the system architecture of an operation management support system. The operation management support system 1 includes at least the operation management support device 2 (of which is a vehicle V, an operation management support device 2, a driving data collection device 15, a prediction result display terminal 31, a biometric data collection device 32, and a business status data collection device 33). The vehicle V, operation management support device 2, driving data collection device 15, prediction result display terminal 31, biometric data collection device 32, and business status data collection device 33 are all capable of communicating via network 100.

[0041] Vehicle V includes: a position sensor 11, a vehicle-to-vehicle distance sensor 12, a speedometer 13, an acceleration sensor 14, and a driving data collection device 15. The position sensor 11, vehicle-to-vehicle distance sensor 12, speedometer 13, and acceleration sensor 14 are examples of sensor equipment (on-board sensors) mounted on vehicle V. Position sensor 11 determines the vehicle's current position using a Global Positioning Satellite System (GNSS). Vehicle-to-vehicle distance sensor 12 detects the distance to vehicles traveling ahead. Speedometer 13 detects the speed of vehicle V. Acceleration sensor 14 detects the acceleration of vehicle V.

[0042] The driving data collection device 15 is mounted on the vehicle V, acquires measurement data from the vehicle V's onboard sensors, and sends it to the operation management assistance device 2. Alternatively, the driving data collection device 15 can be located outside the vehicle V and receive data from the aforementioned onboard sensors via the network 100.

[0043] The prediction result display terminal 31 displays an estimated result of the risk of encountering an accident while driving (hereinafter referred to as accident risk) based on the operation management assistance device 2. Details of the method for estimating the accident risk and the display image of the prediction result will be described later.

[0044] The biological measurement data collection device 32 acquires the driver's measurement data from the biological measuring device and sends it to the operation management auxiliary device 2. Figure 1 The biometric devices used in the example include, for example, a heart rate monitor 34, a thermometer 35, and a blood pressure monitor 36. Drivers or managers use the biometric devices to measure the driver's biological data, for example, before the start of a business day, during driving operations, during non-driving operations, and after the end of a business day.

[0045] The business status data collection device 33 collects the actual business activities performed by the driver as business status data, correlates them with the date and time, and sends it to the operation management auxiliary device 2. The driver can also input the day's business activities into the business status data collection device 33 after the day's work has ended.

[0046] Alternatively, the business status data collection device 33 can also infer the driver's business status process based on the driving data obtained by the driving data collection device 15, and use it as business status data. Alternatively, the business status data collection device 33 can also collect driver behavior data such as driver acceleration, location information, and images of situations captured during business from the biometric data collection device 32, and infer each business status and its process based on the collected behavior data, and use it as business status data.

[0047] Loading 331, unloading 332, and rest 333 are input units, for example, located in the driver's seat of vehicle V, and are buttons operated by the driver. When any one of loading 331, unloading 332, or rest 333 is pressed while vehicle V is stationary, the business status data collection device 33 collects data indicating the pressed state. Furthermore, time information, including the moment the button was pressed, can also be obtained here. Additionally, while vehicle V is in motion, the business status data collection device 33 collects data indicating the movement of vehicle V.

[0048] The operation management auxiliary device 2 maintains multiple programs being processed and the data processed by these programs. The programs executed by the operation management auxiliary device 2 include: hazard determination program 211, estimation model training program 212, biological measurement-business status label association program 213, accident risk estimation program 214, and estimation result prompting program 215.

[0049] The data maintained by the operation management auxiliary device 2 includes: biological measurement data 221, business status data 222, vehicle-mounted sensor data 223, biological measurement-business status data 224, accident risk estimation model 225, and training data 23. Specifically, the training data 23 includes, for example, biological measurement-business status DB231 and hazard assessment DB232.

[0050] Biometric data 221 is the measured biometric data of the driver. Biometric data 221 includes, for example, biometric data measured during operations involving driving, biometric data measured in standby, rest, or other non-operational situations, or a combination of some or all of them.

[0051] Business status data 222 represents the specific business status experienced by the driver during work. Specifically, businesses can include, for example, driving, loading, unloading, waiting, and resting. Furthermore, businesses can be further differentiated based on the type of road traveled, the region traveled, or the type of goods transported, such as driving, loading, and unloading.

[0052] Biometric Measurement-Business Status Data 224 is generated by Biometric Measurement-Business Status Tag Association Program 213 based on Biometric Measurement Data 221 and Business Status Data 222. Biometric Measurement-Business Status Data 224 provides information on what business a driver performed on the same day at the times when biometric data values ​​were measured. Specifically, for example, Biometric Measurement-Business Status Data 224 is obtained by associating Biometric Measurement Data 221 with Business Status Data 222 at various time points.

[0053] The vehicle sensor data 223 is used in the hazard assessment procedure 211. The vehicle sensor data 223 is data received from the driving data collection device 15 related to driving operations, vehicle movement, or location information.

[0054] Accident risk estimation model 225 is a model used by accident risk estimation procedure 214 to estimate the accident risk of a driver encountering an accident using biometric-business status data 224. Accident risk estimation model 225 exists according to the categories of business status labels 502 within the biometric-business status data 224. Furthermore, accident risk estimation model 225 can also exist by time period according to the categories of business status labels 502.

[0055] Training data 23 is used for the inference model training procedure 212. The biometrics-operational status DB 231 within training data 23 contains previously acquired biometrics-operational status data 224. Hazard determination DB 232 contains data acquired through the hazard determination procedure 211 and onboard sensor data 223. This data indicates whether driving in a scenario based on the previously acquired biometrics-operational status data 224 is determined to be hazardous.

[0056] Hazard assessment procedure 211 is a procedure used to determine whether a driver's driving is dangerous. The determination of whether a vehicle is dangerous based on hazard assessment procedure 211 can utilize any technology, and its basis depends on the design. Examples of driving deemed dangerous include sudden braking, sudden acceleration, excessive speed relative to the legal speed limit, and excessively narrow following distances.

[0057] The hazard assessment procedure 211 determines various dangerous driving operations based on vehicle sensor data 223 (measurement data from the vehicle sensors). For example, it determines whether there is sudden braking or sudden acceleration based on the detection data from the acceleration sensor 14, and whether the vehicle is too close to the vehicle based on the detection data from the inter-vehicle distance sensor 12. It also determines whether the vehicle V is adhering to the legal speed limit based on the detection data from the position sensor 11, map information (not shown), and the detection data from the speedometer 13.

[0058] The hazard assessment data in the hazard assessment DB232 can be data representing the hazard level at the same time point as each measurement time point of the biometrics-operational status data 224, or it can be data representing the hazard level at one or more time points several minutes to tens of minutes after the measurement time point of the biometrics-operational status data 224.

[0059] The hazard assessment procedure 211 analyzes the vehicle sensor data 223 sent from the driving data collection device 15 to determine whether each scenario is dangerous. The biometric measurement-service status tag association procedure 213 correlates the biometric measurement data 221 sent from the biometric measurement data collection device 32 with the service status data 222 sent from the service status data collection device 33 to generate biometric measurement-service status data 224. Details of the biometric measurement-service status data 224 will be described later.

[0060] Accident risk estimation procedure 214 estimates the accident risk for the scenario based on biometrics-operational status data 224. Estimation result prompting procedure 215 prompts the predicted accident risk for the driver and its accompanying information. Estimation model training procedure 212 uses biometrics-operational status data 224 and hazard assessment data from hazard assessment procedure 211 to train the estimation model according to the operational status of biometrics-operational status data 224.

[0061] <Example of hardware structure for operation management auxiliary device 2>

[0062] Figure 2 This is a block diagram illustrating an example of the hardware structure of the operation management auxiliary device 2. The operation management auxiliary device 2 includes: a processor 41, a memory 42, an auxiliary storage device 43, an output device 44, an input device 45, and a communication interface (I / F) 46. These components are interconnected via a bus 47. The memory 42, the auxiliary storage device 43, or a combination thereof constitute a storage device for storing data. Figure 1 The program and data shown.

[0063] The memory 42, for example, is a semiconductor memory, primarily used to store executing programs and data. The processor 41 performs various processes according to the programs stored in the memory 42. The processor 41 operates according to the program, thereby realizing various functional units. The auxiliary storage device 43, for example, is a high-capacity storage device such as a hard disk drive or a solid-state drive, used to store programs and data for extended periods.

[0064] The processor 41 can be composed of a single processing unit or multiple processing units, and can contain a single or multiple arithmetic units or multiple processing cores. The processor 41 can be installed as one or more central processing units, microprocessors, microcomputers, microcontrollers, digital signal processors, state machines, logic circuits, graphics processing devices, system-on-a-chip, and / or any device that operates signals according to control instructions.

[0065] The program and data stored in the auxiliary storage device 43 are loaded into the memory 42 upon startup or when necessary. The processor 41 executes the program, thereby performing various processes of the operation management auxiliary device 2. Therefore, the following processes performed by the operation management auxiliary device 2 are based on the processor 41 or the program.

[0066] Input device 45 is a hardware device for allowing the user to input instructions, information, etc., to operation management auxiliary device 2. Output device 44 is a hardware device that displays various images for input and output, such as a display device or printing device. Communication I / F 46 is an interface for connecting to network 100. Input device 45 and output device 44 may also be omitted, and operation management auxiliary device 2 may also be accessed from a terminal via network 100.

[0067] The functions of the operation management auxiliary device 2 can be installed in a computer system consisting of one or more computers, each having one or more processors and one or more storage devices containing non-transient storage media. Multiple computers can communicate via network 100. For example, some of the functions of the operation management auxiliary device 2 can be installed on one computer, while others can be installed on other computers.

[0068] The driving data collection device 15, the prediction result display terminal 31, the biological measurement data collection device 32, and the business status data collection device 33 can each have the same computer structure as the operation management auxiliary device 2. The functions of multiple devices mentioned above can also be installed in one device. For example, the operation management auxiliary device 2, the prediction result display terminal 31, and the biological measurement data collection device 32 can also be integrated into one device.

[0069] <Example of processing performed by operation management auxiliary device 2>

[0070] Figure 3 This is a flowchart of a series of processes in the operation management auxiliary device 2, from the acquisition of biological measurement data 221 to the estimation of accident risk and the presentation of results. First, the operation management auxiliary device 2 acquires various biological data through the biological measurement-business status tag association program 213 and generates biological measurement data 221 (step S101).

[0071] Next, the operation management auxiliary device 2 uses the biometric measurement-business status tag association program 213 to perform data transformation, extraction and shaping on the biometric measurement data 221 for input into the estimation model, associates it with the business status data 222, and generates or updates the biometric measurement-business status DB231 (step S102).

[0072] Next, in order to improve the estimation accuracy, the operation management auxiliary device 2 generates an estimation model based on the business status data 222 associated with the biological measurement data 221 through the estimation model training program 212 (step S103).

[0073] Subsequently, the operation management auxiliary device 2 selects the estimation model corresponding to the biometric-business status data 224 representing the driver's status from the estimation model group of each business status data 222 associated with the biometric data 221 through the accident risk estimation procedure 214, and uses the selected estimation model to estimate the accident risk 241 (step S104).

[0074] Finally, the operation management auxiliary device 2 displays or reports the estimated results (accident risk 241) and the progress of the business status to the driver or their manager through the estimated result prompting procedure 215 (step S105). The details of these processes and the data used for processing are described below.

[0075] <Data Structure Example>

[0076] Figure 4 This is an explanatory diagram illustrating a structural example of biological measurement data 221. In this embodiment, heart rate interval (RR) is used as biological data, and autonomic nervous system function (ANF) index is used as the index of biological measurement data 221. Biological measurement data 221 includes user ID 401, date and time 402, ANF index 403, and body temperature 404 as fields. ANF index 403 includes, for example, total autonomic nervous system power (Total Power) 431 and autonomic nervous system LF / HF 432.

[0077] User ID 401 uniquely identifies each driver as a user. Date and time 402 is the start time of the measurement of biometric data 221 acquired over a certain time span. Here, date and time 402 can also be represented by a measurement end time or a combination of one or more time points that serve as a reference for representing the measurement period, other than the measurement start time.

[0078] The total autonomic power 431 represents the total power (TP), one of the parameters of electrocardiographic variation. Total power is the total power of the power spectrum of a specific frequency band of electrocardiographic variation and is related to fatigue. The autonomic LF / HF 432 is the ratio of the power of the low-frequency band (LF) to the high-frequency band (HF) within a specific frequency band, which is also one of the parameters of electrocardiographic variation. LF / HF represents the overall balance between the sympathetic and parasympathetic nervous systems.

[0079] As another ANF indicator 403, time-domain parameters, frequency-domain parameters, and nonlinear parameters using heart rate or heartbeat RR intervals are considered. Among the time-domain parameters using heart rate or RR, for example, the mean (Mean) and standard deviation (SDNN) of RR intervals measured over a certain time period such as 2 minutes are considered, as well as the coefficient of variation (CVRR), the square root of the mean of the squares of the differences between consecutive adjacent RR intervals (RMSSD), and the total number of consecutive adjacent RR intervals with a difference exceeding 50 ms (NN50).

[0080] In frequency domain parameters using heart rate or RR, in addition to total power and LF / HF, VLF, LF, HF as power spectra in the ultra-low frequency band, or their respective deviations, or CCVVLF, CCVLF, CCVHF as CVRR values ​​of each variable component, are also considered. In nonlinear parameters using heart rate or RR, for example, the standard deviation SD1 of the vertical axis, the standard deviation SD2 of the horizontal axis, and the area of ​​the imaginary ellipse are considered in a Poincaré plot with RR at a certain time point as the horizontal axis and RR at the next time point as the vertical axis. Body temperature 404 is the measured value of the driver's body temperature.

[0081] Figure 5 This is an explanatory diagram illustrating the structure of business status data 222. Business status data 222 includes user ID 401, business status label 502, start date and time 503, and end date and time 504 as fields. Business status label 502 is a label that determines the business status of the driver during work. Specific examples of business statuses determined by business status label 502 include "driving," "loading," "unloading," "standby," and "resting." Alternatively, more detailed information can be stored, such as the type of road traveled, the location of the traveled area, etc., like "driving (general road)," "driving (highway)," and "driving (Kanto)." Furthermore, more detailed information can be stored, such as "road congestion information," "weather information," and "sunshine information," along with the driving conditions. Start date and time 503 and end date and time 504 represent the start and end dates and times of business status label 502 in the same row, respectively.

[0082] Figure 6This is an explanatory diagram illustrating the structure of the Danger Assessment DB232. The Danger Assessment DB232 includes user ID 401, start date and time 602, end date and time 603, and danger assessment result 604 as fields. Start date and time 602 and end date and time 603 represent the dates and times when danger assessment begins and ends for the object data, respectively. Danger assessment result 604 represents the result of the danger assessment procedure 211's estimation of whether the period between start date and time 602 and end date and time 603 constitutes a dangerous scenario. Specifically, it may store, for example, a numerical value indicating the degree of danger, or strings such as "dangerous" or "not dangerous." The format of the danger assessment result 604 corresponds to the format of the output of the danger assessment procedure 211.

[0083] <Acquisition and processing of biometrics - operational status DB231 or biometrics - operational status data 224>

[0084] Figure 7 This is a flowchart illustrating an example of the process for obtaining and processing biological measurement-business status DB231 or biological measurement-business status data 224 based on the biological measurement-business status tag association procedure 213. Figure 7 express Figure 3 The detailed processing example of step S102 shown is as follows. The biometrics-business status label association procedure 213 is executed in two cases: generating the biometrics-business status DB231 used for training to generate or update the inference model, and generating the biometrics-business status data 224 used when applying the inference model to infer accident risk.

[0085] The training data 23 used for training is data obtained from past drivers. The biometrics-operational status DB231 is generated based on the biometrics data 221 and operational status data 222 obtained before the operation of the operation management auxiliary device 2. Furthermore, the training data 23 can be updated with supplementary data, which is extracted periodically from the biometrics data 221 and operational status data 222 obtained after the start of the safe operation system's operation, on a periodic basis ranging from several months to several years. In this case, it is also necessary to store the hazard assessment DB232 corresponding to the supplementary biometrics data 221 and operational status data 222.

[0086] Furthermore, the additional biological measurement data 221, business status data 222, and hazard assessment DB232 (hereinafter referred to as additional data) can also be utilized outside the operation management auxiliary device 2 via the network 100. In this case, the system developer conducts research on whether to add additional data to the training data 23 or exclude conditions from the additional data, and adds the additional data to the training data 23 according to the research results.

[0087] In this embodiment, biological data is defined as RR. First, in the biological data acquisition process (step S171), the biological measurement data collection device 32 measures the driver's RR and sends it to the operation management assistance device 2. Specifically, for example, the biological measurement data collection device 32 measures the driver's electrocardiogram (heartbeat) using a heart rate monitor 34. The biological measurement data collection device 32 detects RR in the measurement results from the heart rate monitor 34. RR represents the interval between peaks of a specific type.

[0088] Figure 8 This is a chart illustrating examples of heart rate fluctuations. Figure 8 The horizontal axis of the chart represents time, and the vertical axis represents electrical potential.

[0089] Figure 9 This is a graph showing the time variation of RR. Return to Figure 7 Next, in the biological measurement data acquisition and processing (step S172), the biological measurement data collection device 32 calculates the ANF index 403 based on RR as biological measurement data 221. Here, as an example, the processing for calculating the total autonomic nerve power 431 and the autonomic nerve LF / HF 432 will be explained.

[0090] Figure 10 This is a graph showing the spectral variation of RR over time. Figure 10 The horizontal axis of the graph represents frequency, and the vertical axis represents the spectral power density of RR. The biometric data collection device 32 calculates LF and HF in the frequency domain, sums them to obtain TP, and divides LF by HF to obtain LF / HF. The biometric data collection device 32 then sends the calculated ANF index 403 as biometric data 221, along with data such as body temperature and blood pressure obtained from other measuring devices, to the operation management auxiliary device 2.

[0091] Next, an example of processing data by extracting biological measurement data 221 according to conditions (step S173) and obtaining biological measurement-business status data 224 or biological measurement-business status DB 231 (step S174) will be described.

[0092] Figure 11This is an explanatory diagram illustrating an example of processing biological measurement-business status data 224 obtained from biological measurement data 221 and business status data 222. In the processing of extracting biological measurement data 221 based on conditions (step S173), the operation management auxiliary device 2 divides the received biological measurement data 221 into a certain time width, such as 2 minutes. The biological measurement data 221 is aligned with a certain time width because the ANF index 403 in the biological measurement data 221 processed in this embodiment is calculated based on an index representing the time change of RR. The certain time width is generally around 30 seconds to 5 minutes.

[0093] Next, the operation management auxiliary device 2 uses basic statistics such as variance and percentage of the segmented biological measurement data 221 to calculate the proportion of improper data contained in the analysis of arrhythmia, measurement error, body movement noise, etc. Only the segmented biological measurement data 221 below a certain threshold is extracted as the extracted biological measurement data 701.

[0094] Next, in the process of obtaining biological measurement-business status data 224 (step S174), the operation management auxiliary device 2 receives business status data 222 from the business status data collection device 33, refers to the business status of the time period corresponding to the extracted biological measurement data 701, and labels the business status of the measurement time point on the extracted biological measurement data 701, and obtains it as biological measurement-business status data 224 or biological measurement-business status DB231.

[0095] Figure 12 This is an explanatory diagram illustrating an example of the structure of the acquired biological measurement-business status data 224. Furthermore, the date and time 402 of the biological measurement data 221 and the start date and time 503 and end date and time 504 of the corresponding business status data 222 may have a certain range of time discrepancies. The labeling of the biological measurement data 221 is for the purpose of identification; therefore, if multiple business status labels 502 exist within the measurement time represented by the date and time 402 included in the extracted biological measurement data 701, the operation management auxiliary device 2 determines which business status label 502 corresponding to the extracted biological measurement data 701 is selected according to a certain rule.

[0096] This rule includes, for example, considering "using a business status label 502 recorded over a longer period in the business status data 222 corresponding to a biological measurement data 221 for a certain time period." Additionally, if the business status label 502 is incomplete, it considers excluding it from the biological measurement-business status data 224 as inappropriate data.

[0097] Figure 13 This indicates the generation or update performed through the inferred model training procedure 212. Figure 3 The flowchart shows a detailed example of the processing procedure for the model of presumed accident risk (step S103). Figure 14 This is an explanatory diagram showing a training example of a presupposition model that uses data divided by business status label 502. The operation management auxiliary device 2 obtains biometrics-business status data 224 from biometrics-business status DB 231 (step S131), and selects business status label 502 from biometrics-business status data 224 (step S132).

[0098] The operation management auxiliary device 2 selects the accident risk estimation model 225 corresponding to the selected business status label 502 (step S225), uses the training data 23 to train the selected accident risk estimation model 225 according to the business status label 502 of biometrics-business status DB231 (step S134), and saves the trained accident risk estimation model 225 (step S135).

[0099] In the presumption model training procedure 212, it is basically assumed that the training data 23 collected in advance is used before the operation of the operation management auxiliary device 2 begins. However, if the training data 23 is updated after the operation of the operation management auxiliary device 2 begins, the accident risk presumption model can be updated by re-executing the presumption model training procedure 212. Furthermore, the coefficients and parameters stored in the accident risk presumption model 225 before the update can also be recorded as a backup inside or outside the system before being overwritten by the update.

[0100] Figure 15 It means Figure 3 The flowchart shown is a detailed example of the accident risk estimation process (step S104) based on the accident risk estimation procedure 214. Figure 16 It means Figure 3 The flowchart shown is a detailed example of the presumption result display processing (step S105). In the presumption result display processing (step S105), the operation management auxiliary device 2 uses the presumed accident risk 241, its corresponding biometric-business status data 224, business status label 502, and a prompt content dictionary 243 that maps the reported and displayed content to generate the presumption result and its accompanying information, and prompts the driver or their manager. Steps S104 and S105 will be described in detail below.

[0101] If the driver is engaged in or attending any business that includes business status label 502, perform accident risk presumption processing (step S104).

[0102] Figure 17This is an explanatory diagram illustrating an example of accident risk estimation processing (step S104). The operation management auxiliary device 2 acquires biometric data 221 and business status data 222 from the biometric data collection device 32 and the business status data collection device 33 in real time, or continuously with a delay of several seconds to tens of minutes.

[0103] In addition, the operation management auxiliary device 2 obtains the biological measurement-business status data 224 generated by the biological measurement-business status tag association program 213 (step S151), selects the business status tag 502 of the biological measurement-business status data 224 (step S152), selects the accident risk estimation model 225 suitable for the selected business status tag 502 (step S152), and performs accident risk estimation (step S134).

[0104] As a presumption, the output biometrics-business status data 224 determines whether the time period of date and time 402 constitutes a dangerous scenario. The presumption can be a binary label such as "dangerous" or "not dangerous," or a continuous value such as "danger level," which represents the probability of an accident occurring or its severity.

[0105] Figure 18 This is an explanatory diagram illustrating an example of feedback prompted during the processing of a presumed "dangerous" situation (step S105). A speaker and other attention-awakening device 1601 mounted on the vehicle V emits a report of biometric data 221 indicating a high risk of accident.

[0106] Additionally, during the reporting process, if the biometrics-service status data 224 input into the estimation model is referenced, and states such as drowsiness, accumulated fatigue, or excitement are estimated based on the biometrics data 221 and service status label 502, the speaker and other attention-awakening devices 1601 mounted on the vehicle V can also issue a call 1603 to confirm whether the driver 1602 has fallen into such a state. Furthermore, the speaker and other attention-awakening devices 1601 mounted on the vehicle V can also issue a call 1604 urging actions that can be performed while performing the service, corresponding to the service status label 502.

[0107] Figure 19This is an explanatory diagram illustrating an example where, after the day's work is completed or the following day, drivers and managers can refer to feedback such as the shift in past accident risks. The feedback screen 1701 on the output device 44 can simultaneously confirm the shift in accident risk and business status over time 1702. Furthermore, for past scenarios recorded as dangerous due to high accident risk, the diagram can display, based on the then-current biometric data 221 and business status label 502, a string 1703 indicating the situation or a candidate string indicating a factor deemed dangerous 1703.

[0108] Thus, according to the above-described embodiment, the operation management assistance device 2 can, for example, even under measurement conditions where multiple business states during driving are irregularly mixed, use biometric data measured from drivers during operations to estimate accident risks corresponding to various business scenarios. Furthermore, the present invention is not limited to drivers of long-distance business vehicles such as logistics trucks and night buses, but can also be applied to short-distance buses, taxis, railways, and airplanes.

[0109] Furthermore, in the above embodiments, the operation management auxiliary system 1 performed... Figure 3 The processing of steps S101 to S105 shown can also be performed as follows: the operation management auxiliary system 1 only performs the learning (generating the presumption model) of steps S101 to S103, and the external system only performs the accident risk estimation of steps S104 and S105. Alternatively, the external system can only perform the learning (generating the presumption model) of steps S101 to S103, and the operation management auxiliary system 1 only performs the accident risk estimation of steps S104 and S105.

[0110] Furthermore, the present invention is not limited to the described embodiments, but includes various modifications and equivalent structures within the scope of the appended claims. For example, the described embodiments are examples that have been explained in detail for the purpose of readily understanding the invention, and the invention is not limited to having all the described structures. Additionally, a portion of the structure of one embodiment may be replaced with the structure of another embodiment. Furthermore, the structure of another embodiment may be added to the structure of one embodiment. Moreover, with respect to a portion of the structure of each embodiment, other structures may be added, deleted, or replaced.

[0111] Furthermore, the aforementioned structures, functions, processing units, and processing modules can be implemented in hardware, such as through integrated circuit design, either partially or entirely, or in software, by having a processor interpret and execute programs that implement each function.

[0112] Information such as programs, tables, and files that perform various functions can be stored in storage devices such as memory, hard disk, SSD (Solid State Drive), or recording media such as IC (Integrated Circuit) card, SD card, and DVD (Digital Versatile Disc).

[0113] Furthermore, the terms "control lines" and "information lines" refer to the lines deemed necessary in the specifications, but are not limited to all control lines and information lines required for installation. In practice, almost all structures can be considered interconnected.

Claims

1. An operation management auxiliary system, comprising: a processor that executes a program; and a storage device that stores the program, characterized in that, The processor can access a first set of associated data that links biometric data related to the driver's organism and business status data related to the driver's business status, and a second set of hazard assessment results representing the hazard of the driver's driving. The processor performs the following processing: The first acquisition process involves obtaining a group of associated data related to the specific business status of the driver from the first set, categorized by business status, and obtaining a group of specific danger assessment results for the specific business status of the driver from the second set. The generation process uses the associated data set related to the specific business state obtained through the first acquisition process and the specific risk assessment result set under the specific business state to generate a presumption model according to the specific business state, and stores it in a third set. The presumption model presumes the accident risk of the driver under the specific business state. The association process links the biometric data with the business status data of the same driver in the same time period and stores them in the first set. as well as The detection process involves identifying illegitimate data representing actions outside the scope of the association process from a time-segmented dataset of biological measurements. In the association process, the processor does not associate improper data detected by the detection process. In the first acquisition process, the processor acquires a group of associated data related to the driver's specific business status from the first set containing associated data linked through the association process. In the association process, the processor performs time segmentation on the biometric data, and for each of the time-segmented biometric data groups, associates the business status data of the same driver in the same time period and stores it in the first set.

2. The operation management auxiliary system according to claim 1, characterized in that, The biological measurement data is based on the driver's heart rate data.

3. The operation management auxiliary system according to claim 2, characterized in that, The data based on the driver's heart rate is related to the driver's autonomic nervous system function.

4. The operation management auxiliary system according to claim 1, characterized in that, The associated data is data that links the biometric data and the business status data of the same driver within the same time period.

5. The operation management auxiliary system according to claim 1, characterized in that, In the detection process, the processor detects the improper data from a group of biological measurement data that is time-divided by heartbeat intervals, when the biological measurement data is based on heartbeat data.

6. The operation management auxiliary system according to claim 1, characterized in that, In the detection process, the processor detects inappropriate data representing behavior based on body movement noise from the time-segmented biological measurement data set.

7. The operation management auxiliary system according to claim 1, characterized in that, The processor performs setting processing to set the business status data based on at least one of the driver's behavior, recorded data of the driver's behavior, and the movement of the vehicle driven by the driver. In the association process, the processor associates the biometric data with the business status data of the same driver in the same time period as set by the setting process, and stores them in the first set.

8. The operation management auxiliary system according to claim 1, characterized in that, The processor performs the following processing: The second step is to obtain the associated data of the predicted object; The selection process involves selecting from the third set a presumption model for the same business state as the business state contained in the associated data of the predicted object obtained through the second acquisition process. The estimation process involves inputting the associated data of the predicted object into the estimation model selected through the selection process, thereby estimating the accident risk for the predicted object. as well as Output processing: Output the estimation result of the estimation process.

9. The operation management auxiliary system according to claim 8, characterized in that, In the output processing, the processor outputs information to remind the driver of the presumed result.

10. The operation management auxiliary system according to claim 8, characterized in that, In the output processing, the processor outputs the prediction result as a displayable result, which includes the shift in the business status contained in the accident risk of the predicted object and the associated data of the predicted object.

11. An operation management assistance method based on an operation management assistance system, the operation management assistance system comprising: a processor that executes a program; and a storage device that stores the program, characterized in that, The processor can access a first set of associated data that links biometric data related to the driver's organism and business status data related to the driver's business status, and a second set of hazard assessment results representing the hazard of the driver's driving. The processor performs the following processing: The first acquisition process involves obtaining a group of associated data related to the specific business status of the driver from the first set, categorized by business status, and obtaining a group of specific danger assessment results for the specific business status of the driver from the second set. The generation process uses the associated data set related to the specific business state obtained through the acquisition process and the specific risk assessment result set under the specific business state to generate a presumption model that estimates the accident risk of the driver under the specific business state, and stores it in a third set. The association process links the biometric data with the business status data of the same driver in the same time period and stores them in the first set. as well as The detection process involves identifying illegitimate data representing actions outside the scope of the association process from a time-segmented dataset of biological measurements. In the association process, the processor does not associate improper data detected by the detection process. In the first acquisition process, the processor acquires a group of associated data related to the driver's specific business status from the first set containing associated data linked through the association process. In the association process, the processor performs time segmentation on the biometric data, and for each of the time-segmented biometric data groups, associates the business status data of the same driver in the same time period and stores it in the first set.

12. A computer-readable storage medium storing instructions, characterized in that, When the instruction is executed by the processor, the processor performs the following steps: The system stores a first set of associated data that links biometric data related to the driver's body and business status data related to the driver's business status, and a second set of hazard assessment results that represent the hazard of the driver's driving. Perform a first acquisition process, acquire a group of associated data related to the driver's specific business status from the first set according to business status, and acquire a group of specific danger assessment results for the driver's specific business status from the second set; The generation process is performed, using the associated data set related to the specific business state obtained through the acquisition process and the specific risk assessment result set under the specific business state, to generate a presumption model that estimates the accident risk of the driver under the specific business state, and stores it in the third set; Perform association processing to associate the biometric data with the business status data of the same driver in the same time period and store them in the first set; as well as The detection process involves extracting inappropriate data from the time-segmented biometric data set that indicates actions outside the scope of the associated processing. In the association process, improper data detected by the detection process is not associated. In the first acquisition process, a group of associated data related to the driver's specific business status is acquired from the first set containing associated data linked through the association process. In the association process, the biometric data is time-segmented. For each of the time-segmented biometric data groups, the business status of the business status data of the same driver in the same time period is associated and stored in the first set.

Images