Simulation device

Abstract

A simulation device according to the present invention comprises: an operator simulation function unit; an operation environment simulation function unit; a simulation function unit; and an observation / analysis function unit. The operator simulation function unit generates, as an operator model, an instance of an agent simulating an operator using the characteristics of the operator as an input. The operation environment simulation function unit generates, as an operation environment model, a GUI stub group simulating an operation environment using an operation log as an input. The simulation function unit activates both models in an operation environment in which both models are virtually arranged, and thereby records the internal state of both models and a simulated operation log. The observation / analysis function unit analyzes the internal state of both models and the simulation operation log, and thereby predicts behavior in real-life operation.

Description

Simulation Equipment

[0001] The present invention relates to a simulation device.

[0002] Simulation is a method for reproducing real-world situations and phenomena on a computer and observing and analyzing their behavior.

[0003] For example, a network simulator places virtual network devices on the software, sets the network topology and various parameters, and simulates them on the software, making it possible to predict various aspects such as the performance of the network when it is actually constructed.

[0004] Furthermore, in recent years, network digital twins, which digitally recreate more realistic networks, have also been proposed, and it is even possible to use network digital twins to predict unknown faults that may occur in actual networks.

[0005] In recent years, research has also been conducted on creating computer-based models known as social simulations, which are used to understand and predict social phenomena, with various applications including people flow simulations.

[0006] Unlike simulations that target networks that are already somewhat systematized from the start, in social simulations, how to represent and model the subject is extremely difficult and important, and various research is being conducted depending on the subject.

[0007] "S4 Simulation System," online, [searched March 11, 2024], Internet: https: / / www.msi.co.jp / solution / s4 / index.html

[0008] On the other hand, there are no established methods for simulating operations.

[0009] In order to improve business efficiency and, in recent years, to achieve digital transformation, various measures are being implemented in operations, such as changing personnel allocation, reviewing business processes, and introducing new systems.

[0010] However, it is extremely difficult to accurately predict the effects of introducing these systems, so in practice, companies often hire skilled consultants from consulting companies to independently estimate the problems with current operations, areas for improvement, and the effects of the improvements based on the consultants' know-how and experience.

[0011] The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a simulation device that enables a precise simulation of operations.

[0012] One aspect of the present invention is a simulation device that simulates operations. The simulation device includes an operator simulation function unit, an operation environment simulation function unit, a simulation function unit, and an observation and analysis function unit. The operator simulation function unit receives operator characteristics as input and generates an instance of an agent that simulates the operator as an operator model. The operation environment simulation function unit receives an operation log as input and generates a group of GUI stubs that simulate the operation environment as an operation environment model. The simulation function unit provides an operation environment creation environment in which an operator model and an operation environment model are virtually arranged, simulates operations by running the operator model and the operation environment model in the operation environment, and records the internal states of the operator model, the internal states of the operation environment model, and the simulated operation log. The observation and analysis function unit predicts behavior in actual operations by analyzing the internal states of the operator model, the internal states of the operation environment model, and the simulated operation log.

[0013] According to the present invention, a simulation device is provided that enables a precise simulation of an operation.

[0014] FIG. 1 is a block diagram showing the functional configuration of a simulation device according to an embodiment. FIG. 2 is a block diagram showing the hardware configuration of a simulation device according to an embodiment. FIG. 3 is a flowchart showing the flow of processing executed by a simulation device according to an embodiment. FIG. 4 is a flowchart showing the flow of operator simulation processing executed by an operator simulation function unit. FIG. 5 is a flowchart showing the flow of operation environment simulation processing executed by an operation environment simulation function unit 12. FIG. 6 is a diagram schematically showing how delays are applied based on actual operations. FIG. 7 is a diagram schematically showing connections between GUI stubs based on processes and how delays are reproduced. FIG. 8 is a diagram schematically showing an example of a communication operation including a chat tool GUI stub and a telephone GUI stub. FIG. 9 is a diagram schematically showing an example of a GUI that provides an environment for creating an operation environment.

[0015] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0016] (Functional Configuration) First, the functional configuration of a simulation device 10 according to an embodiment will be described with reference to Fig. 1. Fig. 1 is a block diagram showing an example of the functional configuration of a simulation device 10 according to an embodiment.

[0017] The simulation device 10 is a device that enables precise simulation of operations. The simulation device 10 has an operator simulation function unit 11, an operation environment simulation function unit 12, a simulation function unit 13, and an observation and analysis function unit 14.

[0018] The operator simulation function unit 11 uses the characteristics of an operator as input and generates an instance of an agent that simulates the operator as an operator model. For example, the agent is RPA (Robotic Process Automation). In the following description, the agent is assumed to be RPA. The operator simulation function unit 11 simulates an operator using RPA software. For example, the operator simulation function unit 11 simulates an operator in units of RPA logic using machine learning. The operator simulation function unit 11 passes the generated operator model (RPA instance) to the simulation function unit 13.

[0019] The operation environment simulation function unit 12 uses the operation log as input and generates a group of GUI stubs that simulate the operation environment as a model of the operation environment. The operation environment simulation function unit 12 passes the generated model of the operation environment (GUI stub group) to the simulation function unit 13. The GUI stub is a software component that provides a virtual GUI that can be automatically operated by RPA. The operation environment simulation function unit 12 also generates processes between the GUI stubs from at least the operation log. The operation environment simulation function unit 12 passes the processes between the generated GUI stubs to the simulation function unit 13.

[0020] The simulation function unit 13 provides an environment for creating an operational environment in which an operator model and an operational environment model are virtually arranged. For example, the environment for creating an operational environment is configured by a GUI that includes the operator model and the operational environment model as GUI components. For example, the GUI enables the creation of an operational environment by arranging and connecting the operator model and the operational environment model by drag and drop.

[0021] The simulation function unit 13 simulates an operation by running an operator model and an operation environment model in the operation environment. More specifically, the simulation function unit 13 simulates an operation by automatically operating the operation environment model using the operator model in the operation environment.

[0022] The simulation function unit 13 records the operation log obtained during the simulation as a simulated operation log. The simulation function unit 13 also records the internal states of the operator model and the operation environment model in chronological order during the simulation. The simulation function unit 13 passes the recorded simulated operation log, the internal states of the operator model, and the internal states of the operation environment model to the observation and analysis function unit 14.

[0023] The observation and analysis function unit 14 predicts behavior in actual operations by analyzing the internal states of the operator model, the internal states of the operation environment model, and the simulated operation log. For example, by analyzing the simulated operation log, the observation and analysis function unit 14 predicts at least the number of order processing, the systems and processes that are bottlenecks, the personnel allocation required to achieve the target processing volume, and the cost-effectiveness of the system to be introduced.

[0024] (Hardware Configuration) Next, a hardware configuration of the simulation device 10 will be described. Here, an example will be described in which the simulation device 10 is configured as a computer. For example, the simulation device 10 may be configured as a personal computer, a server computer, or the like.

[0025] 2 is a block diagram showing an example of a hardware configuration of a simulation apparatus 10 according to an embodiment. The simulation apparatus 10 is configured, for example, by a computer. The simulation apparatus 10 has a control device 20, a display device 40, and an input device 50.

[0026] The control device 20 controls the entire simulation device 10. The control device 20 includes a processor 21, a read only memory (ROM) 22, a random access memory (RAM) 23, and an auxiliary storage device 24.

[0027] The processor 21, ROM 22, RAM 23, auxiliary storage device 24, display device 40, and input device 50 are electrically connected to one another via a bus 30, and can transmit and receive data.

[0028] The processor 21 is configured by a general-purpose hardware processor including, for example, a CPU (Central Processing Unit), a GPU (Graphical Processing Unit), etc. The processor 21 executes programs deployed in the RAM 23 to perform various functions of the simulation device 10.

[0029] The ROM 22 is a non-volatile memory that constitutes part of the main storage device. The ROM 22 non-temporarily stores a startup program required to start up the simulation device 10. The processor 21 loads the startup program in the ROM 22 into the RAM 23 and executes it to start up the simulation device 10. The ROM 22 is, for example, configured from an EPROM (Erasable Programmable Read Only Memory), and is capable of storing various startup settings in addition to the startup program.

[0030] The RAM 23 is a volatile memory that constitutes part of the main storage device. The RAM 23 temporarily stores programs required for processing by the processor 21 and data required for executing the programs. In other words, the RAM 23 functions as a work area for the processor 21.

[0031] The auxiliary storage device 24 is configured with non-volatile memory such as a hard disk drive (HDD) or a solid state drive (SSD). The auxiliary storage device 24 can non-temporarily store various programs executed by the processor 21 and data required for executing the programs. The processor 21 executes various functions of the simulation device 10 by loading the programs in the auxiliary storage device 24 into the RAM 23 and executing them.

[0032] The display device 40 is a device that visually outputs information and displays characters, images, etc. on a screen to provide information to a user. For example, the display device 40 is a liquid crystal display, an organic EL display, a plasma display, etc.

[0033] The input device 50 is a device for a user to input information and instructions, and accepts input of information and instructions. The input device 50 includes a keyboard, a pointing device, etc. The pointing device includes a mouse, a trackpad, a touch screen, etc.

[0034] The display device 40 and the input device 50 may be configured as a device having the functions of both. Such a device may be configured as, for example, a touch panel.

[0035] The input device 50 may also include a device that reads data from a computer-readable recording medium 60 that non-temporarily records data such as a program. For example, the recording medium 60 includes disks such as flexible disks, optical disks (CD-ROM, CD-R, DVD-ROM, DVD-R, etc.), magneto-optical disks (MO, etc.), and semiconductor memories. The input device 50 includes drives and readers for these.

[0036] The program stored in the auxiliary storage device 24 is provided to the simulation device 10, for example, via the recording medium 60. Alternatively, the program may be stored on a server on a network and provided to the simulation device 10 by downloading it.

[0037] For example, when the simulation device 10 is started, the processor 21 executes a startup program in the ROM 22 to start up the operating system (OS). Under the control of the OS, the processor 21 monitors input instructions, connections to external devices, etc. Also, under the control of the OS, the processor 21 sets a program area and a data area in the RAM 23.

[0038] In response to an instruction to start a program, the processor 21 reads the program from the auxiliary storage device 24 into the program area of ​​the RAM 23, and also reads data necessary for executing the program from the auxiliary storage device 24 into the data area of ​​the RAM 23. The processor 21 calculates the data in the data area in accordance with the program and writes the calculation results into the data area.

[0039] Through these operations, the processor 21, RAM 23, and auxiliary storage device 24 work together to execute at least some of the functions of the control device 20. Furthermore, the control device 20, display device 40, and input device 50 work together to execute at least some of the functions of the simulation device 10.

[0040] The input device 50 functions as a functional entity that receives input of operator characteristics. The display device 40 functions as a functional entity that displays the GUI, and the input device 50 functions as a functional entity that performs operations on the GUI. As a result, the control device 20, the display device 40, and the input device 50 enable the creation of an operating environment in which an operator model and an operating environment model are virtually arranged on the GUI.

[0041] The programs non-temporarily stored in the auxiliary storage device 24 include a simulation program that causes the processor 21 to execute at least part of the functions of the control device 20. In other words, the processor 21 executes this simulation program to execute at least part of the functions of the control device 20.

[0042] As a result, the control device 20 works in cooperation with the display device 40 and the input device 50 to execute at least some of the functions of the operator simulation function unit 11, the operation environment simulation function unit 12, the simulation function unit 13, and the observation / analysis function unit 14.

[0043] (Example of Operation) Next, the overall flow of the processing executed by the simulation device 10 will be described with reference to Fig. 3. Fig. 3 is a flowchart showing the overall flow of the processing executed by the simulation device 10.

[0044] In step S11, the simulation device 10 performs operator simulation processing using the operator simulation function unit 11. That is, the operator simulation function unit 11 generates an RPA instance that simulates the operator as a model of the operator based on the input operator characteristics. The operator simulation function unit 11 generates the number of RPA instances required for the simulation. The operator simulation function unit 11 passes the generated RPA instances (operator models) to the simulation function unit 13.

[0045] In step S12, the simulation device 10 performs an operation environment simulation process using the operation environment simulation function unit 12. That is, the operation environment simulation function unit 12 generates a group of GUI stubs that simulate the operation environment as a model of the operation environment based on the input operation log. The operation environment simulation function unit 12 generates the number of GUI stubs required for the simulation. The operation environment simulation function unit 12 passes the generated group of GUI stubs (model of the operation environment) to the simulation function unit 13.

[0046] In step S13, the simulation device 10 performs a simulation process using the simulation function unit 13. For example, the simulation function unit 13 performs an operation simulation by automatically operating an operation environment model (GUI stub group) using an operator model (RPA instance) in the operation environment, and records the internal state of the operator model, the internal state of the operation environment model, and a simulated operation log. The simulation function unit 13 passes the recorded internal state of the operator model, the internal state of the operation environment model, and the simulated operation log to the observation and analysis function unit 14.

[0047] In step S14, the simulation device 10 performs observation and analysis processing using the observation and analysis function unit 14. For example, the observation and analysis function unit 14 predicts behavior in actual operations by analyzing the internal states of the operator model, the internal states of the operation environment model, and the simulated operation log. For example, the observation and analysis function unit 14 analyzes the simulated operation log to predict at least the number of order processing, the systems and processes that are bottlenecks, the personnel allocation required to achieve the target processing volume, and the cost-effectiveness of the system to be introduced.

[0048] (Operator Simulation Process) Next, the operator simulation process executed by the operator simulation function unit 11 will be described with reference to Fig. 4. Fig. 4 is a flowchart showing the flow of the operator simulation process executed by the operator simulation function unit 11.

[0049] The operator simulation process simulates an operator using RPA software. The input is the operator's characteristics, and the output is an RPA instance.

[0050] In step S21, the operator simulation function unit 11 receives the operator's characteristics as an input and acquires teaching data according to the operator's attributes. For example, the teaching data includes the operator's proficiency level.

[0051] In step S22, the operator simulation function unit 11 uses the RPA software and the teaching data acquired in step S21 to learn the logic of the automatic operation of the RPA through machine learning, thereby obtaining an operator model, which is an instance of the RPA.

[0052] In step S23, the operator simulation function unit 11 determines whether the process has been performed for the number of operators required for the simulation.

[0053] If the result of the determination in step S23 is that the processes have not been performed for the required number of operators, the operator simulation function unit 11 repeats the processes of steps S21 to S23.

[0054] If the result of the determination in step S23 is that the processes have been performed for the required number of operators, the operator simulation function unit 11 ends the operator simulation process.

[0055] In relation to operator simulation processing, research is being conducted to incorporate human knowledge and insights into the logic of RPA automatic operations through machine learning. One example of such research is disclosed in Hidetake Koya, Hajime Nakajima, Ken Masuda, and Ikuya Takahashi, "Study on a User Interface Control Method to Prevent Miscontrol," Proceedings of the IEICE General Conference, IEICE, p. 82, March 2013. Another example of research is disclosed in Japanese Patent Application Laid-Open No. 2014-160402.

[0056] All of these methods aim to obtain ideal automated operations (for example, automated operations that do not result in operational errors) as a result of learning, but by deliberately preparing teaching data from operators with low levels of proficiency, it is possible to generate RPA logic that mimics operators with various levels of proficiency.

[0057] For example, operators can be classified by proficiency level, and instructions can be given using the method described above to learn the logic for determining which automatic operations are to be performed for each proficiency level. Instructions can be given directly from the operators, or instruction data can be created by collecting actual operation logs for each proficiency level.

[0058] (Operation Environment Simulation Process) Next, the operation environment simulation process executed by the operation environment simulation function unit 12 will be described with reference to Fig. 5. Fig. 5 is a flowchart showing the flow of the operation environment simulation process executed by the operation environment simulation function unit 12.

[0059] The operation environment simulation process simulates the system, terminal environment, and business processes based on operation logs collected from actual system and terminal operations. The input is the operation log, and the output is a group of GUI stubs.

[0060] In step S31, the operation environment simulation function unit 12 acquires an operation log of an actual operation for a single system.

[0061] In step S32, the operation environment simulation function unit 12 generates a GUI stub that constitutes a model of the operation environment from the operation log acquired in step S31 using machine learning.

[0062] In step S33, the operation environment simulation function unit 12 reproduces the delay between screen transitions. The reproduction of the delay will be described later.

[0063] In step S34, the operation environment simulation function unit 12 determines whether the processing has been performed for the number of systems required for the simulation.

[0064] If the result of the determination in step S34 is that the processes have not been performed for the required number of systems, the operation environment simulation function unit 12 repeats the processes in steps S31 to S34.

[0065] If the result of the determination in step S34 is that processing has been performed for the required number of systems, in step S35 the operation environment simulation function unit 12 acquires an operation log of the actual operation for the entire system including the terminals for each system.

[0066] In step S36, the operation environment simulation function unit 12 generates a process between GUI stubs using machine learning from the operation log acquired in step S35.

[0067] In step S37, the operation environment simulation function unit 12 reproduces the delay between the GUI stubs.

[0068] In relation to operational environment simulation processing, research is being conducted on reproducing business system environments (screens, transitions, and occurring events) from operation logs with the aim of reproducing the environment when creating RPA automatic operation scenarios. One example of such research is disclosed in Hidetake Koya, Hajime Nakajima, Hirotake Nakanishi, and Ikuya Takahashi, "Study on a Method for Automatically Creating GUI Stubs Using Operation Logs to Support the Creation of Control Programs for Automatic Software Operations," IEICE Society Conference Presentation Paper, p. 407, September 2013. Another example of research is disclosed in Japanese Patent Laid-Open Publication No. 2015-43147.

[0069] In addition to these approaches, by adding deep learning and data augmentation techniques, it is possible to reproduce business system environments with greater accuracy.

[0070] In the aforementioned literature, the purpose is achieved if an RPA scenario can be created, so delays between screen transitions and the like are not reproduced.

[0071] However, the simulation device 10 according to the embodiment aims to perform a precise simulation of an operation, and for this purpose, it is necessary to reproduce the operation environment precisely.

[0072] For this reason, the operation environment simulation function unit 12 reproduces the delay between screen transitions in step S33, and reproduces the delay between GUI stubs in step S36. The manner in which a delay is applied based on an actual operation is shown in FIG.

[0073] Typically, in network simulations, the backbone (server) side reproduces the type of network and how systems are connected, but in an operational environment, it is the operators who actually connect the systems, and the operators operate multiple systems to connect them as operations.

[0074] This allows the operator to understand which systems are connected to which systems and how they are being operated by obtaining an overall operation log of system operations and terminal operations.

[0075] For this reason, the operation environment simulation function unit 12 acquires an overall operation log of the actual operation in step S35, generates a process between the GUI stubs from the overall operation log in step S36, and then connects the GUI stubs based on the process and reproduces the delay in step S37. The delay here does not refer to a network delay or the like, but rather to an operational delay. The process-based connection between the GUI stubs and the reproduction of the delay are schematically shown in FIG. 7.

[0076] In recent years, communication between operators is often reproduced using chat tools.

[0077] Therefore, if a chat tool is reproduced as a GUI stub and incorporated into the GUI stub process, communication during operations can also be reproduced.

[0078] In addition, for old-fashioned legacy processes such as telephones that do not have a GUI, they can be reproduced by creating a unique GUI stub that can be operated as a telephone operation from RPA and incorporating it into a process between GUI stubs.

[0079] An example of a communication operation including a chat tool GUI stub and a telephone GUI stub is shown schematically in FIG.

[0080] (Simulation Processing) In the simulation processing, the simulation function unit 13 provides an environment for creating an operational environment in which an operator model and an operational environment model are virtually arranged. For example, the environment for creating an operational environment is configured by a GUI that includes the operator model and the operational environment model as GUI components. For example, the GUI enables the creation of an operational environment by arranging and connecting the operator model and the operational environment model by drag and drop.

[0081] An example of such a GUI is shown schematically in Figure 9. In the GUI shown in Figure 9, operator models can be placed in the placement area by dragging and dropping. Terminal models, which are configured by combining system models (GUI stubs) and processes, can also be placed in the placement area by dragging and dropping. Furthermore, an operation environment can be created by connecting the operator models and terminal models in the placement area by dragging and dropping.

[0082] The simulation function unit 13 performs a simulation of an operation by automatically operating a model of the operation environment using a model of an operator in the operation environment.

[0083] In the simulation, one scenario is selected from the ideal scenarios of the operator's model and executed. The selection probability of the scenario and the calling relationship between scenarios may be defined separately. How the scenario is executed depends on the learned logic.

[0084] When the operator model executes the scenario and automatically operates the operational environment model, events issued by the operational environment are observed. The simulation function unit 13 records the operation log obtained during the simulation as a simulated operation log. The simulation function unit 13 also records the internal states of the operator model and the internal states of the operational environment model in chronological order during the simulation.

[0085] The simulation operation may be performed by constructing a virtual machine for each terminal that performs the simulation.

[0086] (Observation and Analysis Processing) In the observation and analysis processing, the observation and analysis function unit 14 analyzes the internal state of the operator model, the internal state of the operation environment model, and the simulated operation log to predict behavior in actual operations.

[0087] This allows, for example, prediction of the number of orders to be processed in a day in an operating environment where operator models and operating environment models are virtually arranged, prediction of bottleneck systems and processes, prediction of personnel allocation to achieve the target number of orders processed, and prediction of the cost-effectiveness of systems to be introduced.

[0088] Additionally, research has been conducted in recent years into estimating the state of operators using operation logs, and as operator models become more precise and closer to humans, it will become possible to predict the state of operators working in operational environments by analyzing simulated operation logs.

[0089] For example, in recent years, it has been shown that the following indicators can be estimated from operation logs, and simulated operation logs can be applied.

[0090] For example, it has been suggested that stress levels can be estimated by analyzing the relationship between the rate of keystrokes or the amount of operation (such as the number of mouse clicks) and the amount of stress (the amount of salivary amylase secretion). One example of such a suggestion is given in Minako Toba, Takao Sakurai, and Yasuhide Mori, "Correlation Analysis of PC Operation Log Features and Office Worker Stress Levels," Transactions of the Institute of Electronics, Information and Communication Engineers, Vol. J95-D, No. 4, pp. 747-757, January 2016. Another example of a suggestion is given in Minako Toba, Takao Sakurai, Yasuhide Mori, and Masashi Egi, "Multiple Regression Analysis of Office Worker Stress Levels and PC Operation Log Features," Information Processing Society of Japan Digital Practice, Vol. 7, No. 1, pp. 71-79, January 2016.

[0091] It has also been suggested that it may be possible to estimate the level of concentration by analyzing the relationship between the amount of screen change or operation (such as the number of mouse clicks) and the level of concentration based on a subjective assessment. One example of such a suggestion is given in Usami Yuki, Ishizawa Chikako, Kageyama Yoichi, and Shirasu Muneshige, "A Study on the Relationship between Concentration State and Biological Information During PC Work," Proceedings of the Japan Joint Conference on Automatic Control, Vol. 61, pp. 25-28, November 2018. Another example of a suggestion is given in Nagai Yuki, Xu Jianfeng, and Sakazawa Shigeyuki, "Analysis of the Relationship between Subjective Level of Concentration During Deskwork and PC Operation Logs," Proceedings of the Annual Conference of the Institute of Image Information and Television Engineers, Vol. 2014, No. 14-5, pp. 1-2, September 2014.

[0092] It has also been suggested that it may be possible to estimate accuracy by detecting three types of rework from operation logs and integrating their occurrence frequencies. One example of such a suggestion is given in "A Study on Visualizing the Accuracy of System Operations Using Operation Logs" by Hidetake Koya, Akira Kataoka, and Haruo Oishi, ICM, Vol. 122, No. 96, pp. 1-6, July 2022.

[0093] It has also been suggested that it may be possible to evaluate indecision from mouse trajectories based on the detection of conflict in psychology.

[0094] (Effects) As described above, the simulation device 10 according to the embodiment makes it possible to simulate in advance the implementation of various measures, such as changing personnel allocation, reviewing business processes, and introducing new systems, in DX of operations, and to predict the effects, areas for improvement, unexpected issues, etc. In other words, the simulation device 10 according to the embodiment makes it possible to perform a precise simulation of operations.

[0095] In the embodiment, an example has been described in which the simulation device 10 is configured as a computer having at least a processor 21 and a storage device (RAM 23, auxiliary storage device 24), the storage device stores a simulation program, and the processor 21 executes the simulation program to perform the functions of the simulation device 10. However, the simulation program may cause the processor 21 to perform part of the functions of the simulation device 10, that is, the simulation program may cause the processor 21 to perform the functions of the simulation device 10 in combination with a program already stored in the computer. Furthermore, the simulation program may cause the processor to perform the functions of the simulation device 10 in combination with hardware such as a programmable logic device (PLD), a field programmable gate array (FPGA), or a graphic processing unit (GPU).

[0096] The embodiments of the present invention have been described above with reference to the drawings. However, the above embodiment is merely an example of a configuration that embodies the present invention. That is, it is clear that the present invention is not limited to the above embodiment. Therefore, addition, omission, substitution, and other modifications of components may be made within the scope of the technical concept of the present invention.

[0097] In short, the present invention is not limited to the above-described embodiments, and various modifications can be made in the implementation stage without departing from the spirit of the invention. Furthermore, the embodiments may be implemented in appropriate combinations, in which case the combined effects can be obtained. Furthermore, the above-described embodiments include various inventions, and various inventions can be extracted by combining selected elements from the disclosed elements. For example, if the problem can be solved and the desired effect can be obtained even if some elements are deleted from all elements shown in the embodiments, the configuration from which these elements are deleted can be extracted as an invention.

[0098] REFERENCE SIGNS LIST 10: Simulation device 11: Operator simulation function section 12: Operation environment simulation function section 13: Simulation function section 14: Observation and analysis function section 20: Control device 21: Processor 22: ROM 23: Processor 24: Auxiliary storage device 30: Bus 40: Display device 50: Input device 60: Recording medium

Claims

1. A simulation device for simulating operations, comprising: an operator simulation functional unit that takes operator characteristics as input and generates an instance of an agent that simulates the operator as an operator model; an operational environment simulation functional unit that takes operation logs as input and generates a group of GUI stubs that simulate the operational environment as an operational environment model; a simulation functional unit that provides an operational environment creation environment in which the operator model and the operational environment model are virtually arranged, simulates operations by running the operator model and the operational environment model on the operational environment, and records the internal states of the operator model, the internal states of the operational environment model, and the simulated operation log; and an observation and analysis functional unit that predicts behavior in actual operations by analyzing the internal states of the operator model, the internal states of the operational environment model, and the simulated operation log.

2. The simulation device according to claim 1, wherein the operation environment simulation function unit generates, from at least the operation log, a group of GUI stubs that can be automatically operated by the agent, and processes between the GUI stubs.

3. The simulation device according to claim 1, wherein the simulation function unit provides an environment for creating an operational environment using a GUI that enables the creation of an operational environment by arranging and connecting an operator model and an operational environment model by dragging and dropping, simulates operations by automatically operating the operational environment model using the operator model in the operational environment, records the operation log obtained during the simulation as a simulated operation log, and records the internal state of the operator model and the internal state of the operational environment model during the simulation.

4. The simulation device according to claim 1, wherein the observation and analysis function unit analyzes the simulated operation log to at least predict the number of orders to be processed, the systems and processes that are causing bottlenecks, the allocation of personnel to achieve the target number of orders to be processed, and the cost-effectiveness of the system to be introduced.

Images