A digital service market simulation method and system

Abstract

The application provides a digital service market simulation method and system, which models an external demand environment in a digital service market, maps the external demand environment to order states of different types and different difficulties representing user demand, and sets an order arrival strategy and an order difficulty generation strategy; models a subject of a digital service system in the digital service market, the subject including a platform, an organization and an execution agent; models order delivery rules between organizations in the digital service market and execution agent cooperation rules within the organizations; and configures external demand environment, subject and rule parameters of a target digital service market based on modeling results of the three parts, and performs simulation. The application realizes mapping of the digital service market to a computer simulation system through modeling from three aspects of an external environment, a subject and rules, and solves the problem of lack of intuitive and effective simulation methods and experimental systems in the existing digital service market.

Description

Technical Field

[0001] This invention belongs to the field of digital service markets, and specifically relates to a digital service market simulation method and system. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] The digital services market is a new trade model relying on digital technology and the internet. It provides a wide range of services through digital platforms, such as government services, financial services, and communication services. In recent years, the digital services market has expanded rapidly globally, becoming an important engine for economic development and a key to realizing the development of my country's digital economy and comprehensively improving its digitalization level.

[0004] The digital services market consists of digital service systems and the external demand environment. Digital service systems are a new type of service system that utilizes digital technology to provide various services to service demanders, aiming to comprehensively improve service efficiency, quality, and overall service experience, thereby promoting the development of my country's digital economy.

[0005] With the development of technologies such as cloud computing, big data, artificial intelligence, and mobile application development, digital service systems integrate resources, optimize business processes, and utilize advanced technologies to achieve automated, intelligent, and personalized services. In the wave of informatization, digital service systems, with their efficiency, convenience, and accuracy, have become important tools for modern enterprises and individuals, providing higher-quality, more efficient, and personalized services to service requesters and playing a vital role in various fields. For example, intelligent customer service systems provide 24 / 7, personalized service experiences; data service systems, through data analysis, help enterprises identify problems in the service process and optimize service workflows.

[0006] A digital service system is an ecosystem in which participants achieve efficient collaboration by establishing a series of interactive relationships (such as social relationships, service relationships, and supply relationships). For example... Figure 1 As shown, the service operator employs multiple supply-demand matching strategies, integrating the business services of different service providers to offer candidate service solutions to customers. Based on customer feedback, the operator continuously adjusts its corporate relationship network and the operational status of physical equipment, thereby influencing the performance of the service matching strategies and ultimately forming a closed-loop digital service system.

[0007] Monitoring and analyzing the complex dynamics and operational status among stakeholders helps guide service matching between supply and demand sides and improves multi-party collaboration efficiency. As the digital service market increasingly exhibits characteristics of social, physical, and information integration, effective system governance and ensuring service stability and efficiency are crucial. However, digital service systems often involve human and social factors; system complexity and the variability of the external environment make effective system evaluation difficult and system anomalies frequent, severely impacting service continuity and the user experience. Summary of the Invention

[0008] To address the aforementioned problems, this invention proposes a digital service market simulation method and system. This invention models the digital service market into a computer simulation system from three aspects: external environment, subjects, and rules, thus solving the problem of the lack of intuitive and effective simulation methods and experimental systems in existing digital service markets.

[0009] According to some embodiments, the present invention adopts the following technical solution:

[0010] A digital services market simulation method includes the following steps:

[0011] Model the external demand environment in the digital service market, map the external demand environment to different types and difficulties of order status representing user needs, and set order arrival strategies and order difficulty generation strategies.

[0012] The entities of digital service systems in the digital service market are modeled, including platforms, institutions, and executing intelligent agents, forming an entity class table for each entity;

[0013] Model the order transmission rules between institutions in the digital services market and the collaboration rules of execution agents within institutions;

[0014] Based on the modeling results of the three parts, the external demand environment, subjects, and rule parameters of the target digital service market are configured, and simulation is performed to obtain simulation results.

[0015] As an alternative implementation, the process of mapping the external demand environment to order states representing different types and difficulties of user demand includes constructing an order entity attribute table, which includes the order type, name, pending queue, order value to order difficulty ratio, the agent accepting the order, order status, order arrival time, order delivery connection, and upper and lower limits of the order difficulty threshold.

[0016] As an alternative implementation, the order type is set with multiple basic orders, each of which is matched with a corresponding institution in the digital service system, and each order has a corresponding difficulty value.

[0017] As an alternative implementation, the process of setting an order arrival strategy includes: generating order arrival times based on queuing theory and setting an order arrival rate that can be adjusted by the user to control the average number of orders arriving per unit of time.

[0018] As an optional implementation method, the process of setting the order difficulty generation strategy includes: the distribution of orders provided by the user includes normal distribution and uniform distribution. In the case of normal distribution, multiple difficulty generation levels are provided to the user based on the length of the order and the overall difficulty, and the order probability of each level is controlled according to the normal distribution point value.

[0019] In a uniform distribution scenario, users cannot select the difficulty level of the generated order, and the types and difficulty values ​​of generated orders are uniformly distributed.

[0020] As an alternative implementation method, the process of modeling the entities of digital service systems in the digital service market includes:

[0021] The platform is used to coordinate and control the entire digital service system, receive orders from outside the digital service system, allocate orders to different institutions for processing, and calculate total costs and total revenue to evaluate overall efficiency.

[0022] The organization consists of internal salespersons and software robots, used to receive orders assigned by the platform and to transfer orders between organizations;

[0023] The executing intelligent agent includes a salesperson and a software robot. The salesperson is configured to handle orders of all difficulty, and the salesperson can create value by completing the order. The value created varies depending on the difficulty, processing time, and quantity of the order.

[0024] The software robot is configured to handle orders with a difficulty level lower than a set threshold, but its order processing speed is higher than that of a salesperson, and the software robot does not incur additional costs when processing orders.

[0025] As an alternative implementation method, the process of modeling the order transmission rules between institutions in the digital service market and the collaborative rules of the execution agents within the institutions includes: the platform is responsible for receiving orders from external sources, allocating them to institutions according to a set algorithm, and the overall efficiency of the platform reflects the quality of the service.

[0026] Different orders are processed in parallel, while the same order is processed serially.

[0027] After receiving an order from a platform or other organization, the organization selects an execution agent to process the order according to its internal order allocation strategy. The execution agent has the right to decide whether to establish a connection with other execution agents to jointly process the order.

[0028] Each executing agent forms a collaborative network. Only executing agents that are connected within the collaborative network can establish connections to jointly process orders. Moreover, only agents with the attribute of "salesperson" can proactively initiate connections to other executing agents.

[0029] As a further implementation method, the process of allocating orders to institutions according to a set algorithm includes: the platform determines the institutions participating in the allocation based on the type of the order; sorts the completion order of the institutions participating in the allocation according to the difficulty of the order; and selects the institution with the lowest occupancy rate to prioritize the allocation of the order; if all software robots of the currently allocated institution are already occupied, the order is prioritized to the institution whose software robots are not fully occupied according to the allocation order.

[0030] As a further implementation, the process of selecting the execution agent to handle the order according to the internal order allocation strategy of the organization includes: if the difficulty of the order is less than the difficulty threshold that the software robot can handle, then the idle software robot is used first to handle it; if all software robots are occupied, then the execution agent with the fewest unprocessed tasks in the organization is queried to handle the order.

[0031] If the difficulty of an order exceeds the threshold that the software robot can handle, the system will query the salesperson with the fewest unprocessed tasks within the organization and assign the order to that salesperson for processing.

[0032] As a further implementation, after each order is assigned by the organization, the status of each execution agent within the organization is updated, including the occupancy rate and the total occupancy rate of the organization. After the execution agent within the organization finishes processing the order, the execution agent returns the order to the organization layer, which then judges and passes it to the next organization to process the order.

[0033] As a further implementation method, after the various executing agents form a cooperative network, the rules for processing orders include:

[0034] If the order difficulty value is less than the difficulty threshold that the software robot can handle, the salesperson will choose other idle execution agents that are connected to it to cooperate in processing;

[0035] If the difficulty of an order exceeds the threshold that the software robot can handle, the salesperson will choose to collaborate with other available salespersons who are connected to them.

[0036] If other execution agents that have established connections with this execution agent are all in an occupied state, then this execution agent can only process the order alone.

[0037] As a further implementation method, when the intelligent agent processes an order, it is necessary to calculate the order processing time based on the difficulty value of the order and the working efficiency of the corresponding intelligent agent. If the difficulty value of the order and the working efficiency of the corresponding intelligent agent are not divisible, the order processing time is calculated by rounding up.

[0038] If multiple agents collaborate to process an order, the order processing time is calculated based on the order's difficulty level and the sum of the work efficiencies of each participating agent.

[0039] As a further implementation, after the intelligent agent processes the order, the efficiency of the digital service system is calculated as the difference between revenue and cost.

[0040] A digital services market simulation system, comprising:

[0041] The external demand environment module is configured to model the external demand environment in the digital service market, map the external demand environment to different types and difficulties of order status representing user needs, and set order arrival strategies and order difficulty generation strategies.

[0042] The entity modeling module is configured to model the entities of digital service systems in the digital service market, which include platforms, institutions and executive agents, forming an entity class table for each entity;

[0043] The rule modeling module is configured to model the order transfer rules between institutions in the digital services market and the collaborative rules of execution agents within institutions.

[0044] The simulation execution module is configured to perform simulations based on the modeling results of three parts: the external demand environment, subjects, and rule parameters of the target digital service market, and to obtain simulation results.

[0045] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0046] This invention can effectively simulate digital service market scenarios. From a modeling perspective, it achieves a mapping from the digital service market to a computer simulation system by modeling the external environment, the subject, and the rules. Functionally, the environment design allows for personalized customization of order parameters, and the subject design supports custom attributes and validation.

[0047] This invention can intuitively display order transmission and agent working status. By changing the system environment from multiple dimensions and adjusting the main parameters within the system, it helps system administrators observe the parameters affecting system performance from multiple perspectives, aiming to obtain the maximum benefit under actual conditions.

[0048] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0049] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0050] Figure 1 This is a logic diagram of the digital service market operation in Example 1;

[0051] Figure 2 This is a business process diagram of the digital service system in Implementation Example 1;

[0052] Figure 3 This is the Agent operation association diagram in Example 1;

[0053] Figure 4 This is an example of a collaborative network diagram within an organization, as shown in Example 2.

[0054] Figure 5 This is a diagram of the inter-organizational order transfer network in Example 2;

[0055] Figure 6 This is a diagram illustrating the agent collaboration network and data within the organization in Example 2. Detailed Implementation

[0056] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0057] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0058] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0059] Where there is no conflict, the embodiments and features described in this application may be combined with each other.

[0060] Example 1

[0061] This embodiment discloses a modeling method for the digital services market, which specifically includes the following:

[0062] Step 1: External Environment Modeling

[0063] In the digital services market, the external environment of a digital service system is its external demand, represented by the continuous influx of orders from different users. The time interval, quantity, type, and complexity of the arriving orders must all be considered. The time it takes to complete an order reflects the system's ability to fulfill orders, while the time and cost incurred by the system in completing all incoming orders measures the system's overall capability.

[0064] Orders originate from outside the system and are categorized into N basic types, such as ABC..., and their random combinations. A BA type order is equivalent to an AB type order, and so on. Furthermore, orders of the same type also have varying levels of complexity (i.e., processing difficulty). Specific order entity attribute settings are shown in Table 1.

[0065] It should be noted that the examples shown in the tables below are all examples. Changes may be made in other embodiments.

[0066] Table 1 Order Entity Attribute Table

[0067]

[0068] The specific order environment operation mechanism is designed as follows:

[0069] (1) Order type design

[0070] Based on the order settings, there are N basic order types, such as ABC... where N is a positive integer. A complex order can be composed of multiple basic order types, such as "Order ABH", "Order CEF", "Order ACDEFH", etc. The digital service system correspondingly sets up N types of agencies of the same type, and agencies of the same type can only handle orders of the same type. For example, agency A can only handle orders of basic type A. Orders should also have different difficulty values, representing the difficulty of the order. The value is a positive integer in the range [a, b], and salespeople and software robots can handle orders of different difficulties. At the same time, the difficulty value of an order also indirectly represents the value of the order. For example, if an agency handles an order with a difficulty of 100, then it can obtain... , of which This is a system-defined ratio representing the mapping between order difficulty and the value created by the Agent in processing the order. Therefore, an order can ultimately be represented as "Order A78B29E52". In addition, the system internally specifies the maximum order difficulty that the robot can handle. .

[0071] (2) Order arrival strategy

[0072] To simulate the arrival of user demand in real-world scenarios, the order arrival time will be generated based on queuing theory. In addition, to adjust the number of orders arriving in the system, the system has set an "order arrival rate" that users can adjust to control the average number of orders arriving in the system per unit of time.

[0073] (3) Order difficulty generation strategy

[0074] The difficulty of orders is a characteristic of the system's external environment and is one of the factors that need to be considered. The distribution of user-submitted orders includes normal distribution and uniform distribution.

[0075] Normal distribution. Under a normal distribution, the system provides users with three difficulty levels to choose from: low, medium, and high. Low-difficulty orders are shorter and have a lower overall difficulty level, while high-difficulty orders are longer and have a higher difficulty value. The system will control the probability of generating "low-difficulty," "medium-difficulty," and "high-difficulty" orders according to the normal distribution values. , and .

[0076] Uniform distribution. In a uniform distribution scenario, users cannot select the difficulty level for generating orders; the system will generate orders of both type and difficulty level evenly.

[0077] Step 2: Main Body Modeling

[0078] The digital service system involves four types of entities: "platform", "organization", "salesperson" and "software robot", which are modeled as follows.

[0079] (1) Platform

[0080] The platform is responsible for coordinating and controlling the entire system, receiving orders from outside the system, allocating orders to different institutions for processing, calculating total costs and total revenue, and evaluating overall efficiency. Specific platform entity attributes are shown in Table 2.

[0081] Table 2 Platform Entity Classification Table

[0082]

[0083] (2) Institution

[0084] The organization consists of internal salespersons and software bots, responsible for receiving orders assigned by the platform, and orders can be transferred between organizations. The organization class inherits from threading.Thread, supporting multi-threaded operation. Specific organization entity attributes are shown in Table 3.

[0085] Table 3 Institutional Entity Classification Table

[0086]

[0087] (3) Salesperson

[0088] Salespeople are one of the main entities handling orders within the organization. Completing orders creates value, which varies depending on the difficulty, processing time, and quantity of the order. Based on the value created, the salesperson's performance will increase accordingly. Unlike robots, salespeople can handle orders of all difficulty levels; however, generally speaking, their processing efficiency is lower than that of robots. Specific salesperson entity attributes are shown in Table 4.

[0089] Table 4 Salesperson Entity Class Table

[0090]

[0091] (4) Software robots

[0092] Software robots (hereinafter referred to as robots) are another component of the organization's order processing mechanism. Compared to salespeople, robots can only handle low-difficulty orders, but they process orders much faster and do not incur additional costs. Specific attributes of the software robot entity are shown in Table 5.

[0093] Table 5. Software Robot Entity Classification Table

[0094]

[0095] Step 3: Rule Modeling

[0096] like Figure 2 As shown, the system supply side is designed with a three-layer structure of "platform-institution-Agent". The platform is responsible for receiving orders from outside the system and allocating them to institutions according to a certain algorithm. At the same time, the overall efficiency of the platform can reflect the quality of the system service.

[0097] Different orders are processed in parallel, while the same order is processed serially. For example, for an order of type ABCD, only after organization A has finished processing part A will it be handed over to organizations B, C, and D to process the remaining part; of course, organization B can also process part B first, and then organizations A, C, and D can process the remaining part.

[0098] Orders can be passed between institutions. For example, "Order ABC" might be processed as follows: Institution A processes it first, then passes it to Institution C for processing, and finally Institution B processes it. After receiving an order from the platform or other institutions, an institution selects an Agent to process the order according to its internal order allocation strategy. This Agent has the right to decide whether to establish a connection with other Agents to jointly process the order.

[0099] Each organization has multiple agents, divided into salespeople and bots. These agents form a collaborative network, and only agents connected within the network can establish a connection to jointly process orders. During system operation, only salespeople can proactively initiate connections to other agents. The system defines that an order can be processed collaboratively by more than two agents; multiple agents establishing connections to process the same order is also allowed. The system operates on two collaborative relationships: "human-machine collaboration" and "human-human collaboration."

[0100] (1) Inter-institutional order allocation strategy

[0101] The platform prioritizes assigning orders to institutions based on their occupancy rates. For example, if an order of type ABCD arrives at a certain moment, with the difficulty value of each component adding up to a complete order A12B20C25D80, and institution A has a occupancy rate of 30%, institution B 50%, institution C 40%, and institution D 80%, the platform will assign the order to institution A. After institution A completes its order, if there are remaining orders B, C, and D, institution A will then decide which institution to pass the order on to based on the current occupancy rates of institutions B, C, and D.

[0102] Assuming organization A processes all orders, organization B has the lowest occupancy rate at this point, but it's possible that all robots in organization B are currently occupied. Although robots can only handle lower-difficulty orders, they don't incur the additional costs of human salespeople. Currently, there are unprocessed orders B20C25D80. These orders, regardless of whether they are assigned to organization B or organization C, will be prioritized for processing by their respective robots. Even if organization C has a higher overall occupancy rate than organization B, there might still be idle robots within organization C. Order allocation strategies need to comprehensively consider order difficulty, internal robot occupancy rates, and human salesperson occupancy rates.

[0103] This embodiment takes order "A12B40C20D87" as an example, and the specific process is as follows:

[0104] When processing orders, prioritize the parts with difficulty level 1, and the processing order is A12C20→B40D87;

[0105] When the platform wants to process part A12C20, it will compare the robot occupancy rates within mechanisms A and C, and prioritize the mechanism with the lower occupancy rate.

[0106] If all robots in organizations A and C are currently in use, then compare the usage ratio of salespersons in organizations A and C.

[0107] In extreme cases, the utilization rate of robots and salespersons in organizations A and C is 100%. Only then should we consider part B40D87 to alleviate the pressure on organizations A and C.

[0108] In an even more extreme scenario, if all four institutions (A, B, C, and D) are at 100% occupancy, the order will be assigned to the institution with the least "pressure" and await processing by the assigned agent.

[0109] (2) Inter-institutional order allocation strategy

[0110] Once the order arrives at the agency, the agency will first obtain the portion of the order that requires its own processing. The specific processing procedure is as follows:

[0111] If the difficulty level is below the threshold that the robot can handle, an idle robot will be used first. If all robots are occupied, the most "idle" agent (with the fewest unprocessed tasks) within the organization will be selected to handle the order.

[0112] If an order's difficulty exceeds the threshold that the robot can handle, the system will query the organization's most "idle" salesperson and assign the order to that salesperson for processing.

[0113] Each time an organization assigns an order, it needs to update some statuses of the agents within the organization (Agent utilization rate, total organization utilization rate, etc.). After the agents within the organization have finished processing the order, the agents will return the order to the organization layer, which will then judge and pass it on to other organizations that will process the order.

[0114] (3) Human-computer collaboration / human-human collaboration strategy

[0115] Initially, network connections are generated between agents within the organization based on user selections. For example, in the G(N,P) random network mode, it is assumed that there is a 1 / 4 probability that any two agent nodes can connect (dashed line connection). If 12 agents are initially preset, then the expected number is... One connection. When a salesperson receives an order, the processing rules are as follows:

[0116] If the order difficulty value is less than the difficulty threshold that the robot can handle, the salesperson can choose to collaborate with other idle agents (including robots and salespeople) that are connected to it.

[0117] If the difficulty of an order exceeds the threshold that the robot can handle, the salesperson can choose to collaborate with other available salespersons who are connected to them.

[0118] When the remaining amount of an order is less than the difficulty threshold that the robot can handle, the salesperson can choose to collaborate with an idle robot to process the order.

[0119] If all other agents connected to this agent are occupied, this agent can only process the order alone. Of course, during the processing, the agent will also continuously monitor whether other agents connected to it are idle, and if they are idle, it may establish a connection with them.

[0120] Agent collaboration can speed up processing, and two-person collaboration can even be faster than robot processing. However, robots do not incur additional costs for processing orders. Furthermore, if the collaboration is too complex, it may even increase the waiting time for subsequent orders.

[0121] Once the order is processed by one agency, it will be passed on to the next agency for processing.

[0122] (4) Rules for calculating order processing time

[0123] When an order is generated, there will be a difficulty level. Orders with higher difficulty can only be handled by sales staff. The difficulty level corresponds to the amount of work required to complete the order.

[0124] Every Agent has a performance factor (eff) as a property, where eff is a positive integer. Initially, the system would base its performance on the average performance of the robots selected by the user. robot_avg_efficacy Average work efficiency of salespersons people_avg_efficacy The robot's work efficiency is generated using a normal distribution, while the salesperson's work efficiency is generated using a uniform distribution.

[0125] When the Agent processes an order, assuming the order difficulty is 35 and the Agent's efficiency is 5, then the Agent needs 7 units of time to process this order. When the order difficulty is 32, the Agent also needs 7 units of time to process this problem. In order to facilitate the timing operation of the system, the Agent will not accept other orders during the remaining 0.6 units of time.

[0126] When agents collaborate to process orders, if Agent1 and Agent2 are collaborating on an order with a difficulty level of 90, Agent1's efficiency is 10, and Agent2's efficiency is 6, then the time it takes for them to collaboratively process this order is... Unit of time.

[0127] System server architecture as follows Figure 3 As shown, when processing orders, each organization is set to occupy one thread, and the agents within the organization are set to run serially. With each thread running at a relatively fast speed, it is possible to simulate the situation where all agents run in parallel.

[0128] (5) System performance calculation rules

[0129] System performance reflects the operational capabilities of a digital service system. To align with real-world scenarios, performance in digital service systems is defined as revenue minus costs. Furthermore, to observe system performance in detail, it's crucial to consider factors beyond the institutional and platform levels; the value created by agents within an organization is the primary source of both system costs and revenue.

[0130] Robots require maintenance costs for each cycle, but they can also create value by processing orders. The calculation formula is:

[0131] ;

[0132] In the formula, The number of orders processed by the software robot; This represents the total number of orders processed by all agents involved in order processing. This represents the total value that can be generated by processing orders of this difficulty.

[0133] For sales representatives, there is a fixed base salary each cycle, plus performance-based pay based on their work. These two items are the platform's main costs. Sales representatives also create value for the platform by processing orders. The calculation formula is:

[0134] ;

[0135] In the formula, The number of orders processed on behalf of the salesperson; This represents the total number of orders processed by all agents involved in order processing. This represents the total value that can be generated by processing orders of this difficulty.

[0136] Business employee wages The calculation formula is:

[0137] ;

[0138] In the formula, This represents the basic salary earned by a salesperson without a fixed sales cycle. This represents the ratio by which the value of an order of corresponding difficulty is converted into the salesperson's earnings.

[0139] For the organization, costs include the maintenance fees for robots per cycle and the salaries of sales staff. Revenue is the total value created by all its robots and sales staff multiplied by the ratio of the organization's total value created to its own revenue. This ratio is compared to the ratio used when calculating sales staff salaries. value_ratio It needs to be high. Under normal circumstances, the organization processes all orders, so the total value created by all robots and salespeople within the organization equals the sum of the difficulty of all orders received by the organization. The organization's efficiency is the difference between its revenue and costs. This can be formally expressed as follows:

[0140] Institutional costs The calculation formula is:

[0141] ;

[0142] In the formula, This represents the maintenance cost of the robot per cycle; This represents the salesperson's salary for each cycle.

[0143] Institutional Revenue The calculation formula is:

[0144] ;

[0145] In the formula, The total value created by all robots and salespeople within the organization for processing orders; The ratio representing the mapping between value creation and the organization's own revenue, with a value greater than [value missing]. .

[0146] Institutional effectiveness The calculation formula is:

[0147] ;

[0148] Calculating system performance from the platform's perspective helps system operators adjust system parameters. After a salesperson completes a task, the platform updates their performance and some of the organization's data. Based on system evaluation metrics (calculated by combining task difficulty, number of tasks, and customer satisfaction), the system aims to ultimately achieve self-regulation.

[0149] Platform Costs The calculation formula is:

[0150] ;

[0151] Platform Revenue The calculation formula is:

[0152] ;

[0153] In the formula, The total value created by processing orders for all institutions within the system; The ratio representing the mapping between the platform's value and its own revenue, with a value greater than [missing value]. ; Customer satisfaction with completed orders.

[0154] Customer satisfaction The calculation formula is:

[0155] ;

[0156] In the formula, This refers to the actual processing time of the order. The customer's expected processing time for the order; This is the average processing time for orders of similar difficulty in the past; Customer psychological volatility represents the different psychological states of different customers regarding the order completion time.

[0157] System performance The calculation formula is:

[0158] .

[0159] Example 2

[0160] The purpose of this embodiment is to provide a digital service marketplace simulation system developed based on a full-stack Python architecture. The goal of this digital service system simulation is to create a user-friendly graphical system that is simple, clear, and aesthetically pleasing, while also enabling real-time data transmission and display. Therefore, this system adopts a Python + Vue front-end / back-end separation framework. The Python back-end runs on a cloud server with its program ports exposed to the outside world. The front-end runs locally for easy adjustments during development. Since the system involves many data chart displays, the front-end integrates the Echarts framework, and Element-Plus is used to optimize the front-end pages.

[0161] The aforementioned digital service market simulation system includes:

[0162] A digital services market simulation system, comprising:

[0163] The external demand environment module is configured to model the external demand environment in the digital service market, map the external demand environment to different types and difficulties of order status representing user needs, and set order arrival strategies and order difficulty generation strategies.

[0164] The entity modeling module is configured to model the entities of digital service systems in the digital service market, which include platforms, institutions and executive agents, forming an entity class table for each entity;

[0165] The rule modeling module is configured to model the order transfer rules between institutions in the digital services market and the collaborative rules of execution agents within institutions.

[0166] The simulation execution module is configured to perform simulations based on the modeling results of three parts: the external demand environment, subjects, and rule parameters of the target digital service market, and to obtain simulation results.

[0167] Building and running a digital services marketplace simulation system mainly involves the following steps:

[0168] Step 1: Environment Setup

[0169] In the environment configuration module, data requiring real-time display, such as "Order Transmission" and "System Performance," utilizes WebSocket technology for real-time data interaction between the front-end and back-end. User-initiated one-way access requests to the system are handled using the Flask framework for this interaction. The system order environment parameters are configurable as follows:

[0170] Order generation method: Use the selector and set the options to "normal distribution" or "uniform distribution".

[0171] Order difficulty: Use a numeric input box, set it to only accept positive integers, including minimum and maximum values.

[0172] Order arrival rate: This represents the average number of orders arriving per unit of time. You can set an integer or decimal range, such as [0.5, 10].

[0173] Period duration: Use a numeric input box, set to only accept positive integers, including minimum and maximum values.

[0174] Environment design parameters are submitted to the backend's " / envDesign" interface via a form. The form needs to be validated by the frontend rules before submission.

[0175] Step 2: Main Body Settings

[0176] The main parameters that can be set in the system are as follows:

[0177] Number of agents within the organization: includes minimum, maximum and step size.

[0178] Human-machine ratio: includes minimum value, maximum value and step size.

[0179] Robot cycle maintenance cost: includes minimum, maximum and step size.

[0180] Salesperson base salary: includes minimum, maximum and step size.

[0181] Salesperson performance: includes minimum value, maximum value, and step size.

[0182] Robot average work efficiency: includes minimum, maximum and step size.

[0183] Average salesperson work efficiency: includes minimum, maximum and step size.

[0184] Institutional network type: Use a selector for users to choose from.

[0185] Most of the system's main parameters are numerical, so a slider combined with data is used for display, which facilitates adjustments and reduces input errors.

[0186] The main design parameters are submitted to the backend " / agentDesign" interface via a form. The form needs to be validated by the frontend rules before submission.

[0187] After setting all environment and subject parameters and submitting, users can preview the generated Agent. By clicking the "Agent Preview" button on the "Subject Design" interface, users can access the backend's " / getAgent" API and enter the Agent details page. This page displays all the Agents generated in a table format, including each Agent's name, type, affiliated organization, work efficiency, robot's maintenance cost per cycle, salesperson's base salary per cycle, and the connection relationships between Agents. This page allows users to locate Agents using names through fuzzy searches; for example, a search for "C" will return results like "C_Robot_1, C_Robot_2, …, C_People4". Users can also filter by Agent type (salesperson or robot).

[0188] Step 3: System Operation

[0189] The system's backend logic comprises two main modules: order transmission and human-machine collaboration, as detailed below:

[0190] (1) Order transmission

[0191] After initial parameter settings are complete, users can click to enter the running interface. This interface is implemented using a relationship diagram in Echarts, containing N institutional nodes. The running interface includes functions for starting, pausing, and resuming the system. The main system interface also allows users to view platform efficiency, institutional costs and revenues, and the number of different types of orders arriving in the system in real time.

[0192] (2) Human-machine collaboration

[0193] To allow users to intuitively query the activities and data of agents within an organization, the interface also allows users to click on the organization to enter its internal network diagram. In the network diagram, dashed lines between agents indicate that two agents can establish a collaborative relationship, while solid lines indicate that agents are collaborating to process the same order. Simultaneously, the internal network diagram displays the organization's real-time costs, revenues, internal network type, and agent occupancy rate, all presented using a unified description list from Element-Plus. Furthermore, while users observe the network diagram, the interface displays the organization's real-time performance, average agent performance, robot performance, and salesperson performance; the first two are presented using line charts, while the latter two are displayed using bar charts.

[0194] The organization's internal network can be configured in various types, such as Figure 4 As shown, the preset networks used for human-computer collaboration include 5 types of regular networks, 2 types of random networks, 2 types of small-world networks, and 1 type of scale-free network. These networks will now be introduced.

[0195] Isolated node network: Each agent in the network has no direct connection with other agents, cannot form a cooperative relationship with other agents, and can only process orders independently.

[0196] Globally Coupled Network: In this network, each agent has relationships with other agents and can form cooperative relationships with any agent. The network type is a complete graph.

[0197] Star network: In a star network, there is a key node. Agents located at this node have relationships with other agents, but other agents can only form cooperative relationships with agents located at the key node and cannot cooperate with other agents.

[0198] One-dimensional ring: Agents form a ring network, and each agent can only form a cooperative relationship with two adjacent agents.

[0199] Two-dimensional lattice: In a two-dimensional lattice network, an agent can form a cooperative relationship with agents in the four directions of "up, down, left, and right". The two-dimensional lattice network requires the number of agents to be a perfect square number.

[0200] A G(N,L) random network consists of N agent nodes connected to each other via L randomly placed links. Initially, L is set to ;

[0201] A G(N,P) random network consists of N nodes, where each pair of nodes is connected with a probability of P, where P is set to 0.25. .

[0202] WS Small-World Network: Consider a nearest-neighbor coupled network of N nodes forming a loop, where each node is connected to its left and right neighbors (K / 2 nodes in total), where K is an even number. The parameters satisfy N >> K >> ln(N) >> 1 (K-nearest-neighbor regular network). Then, each edge in the network is randomly reconnected with probability p, meaning one endpoint of the edge remains unchanged while the other endpoint is replaced with a randomly selected node in the network. Small-world networks stipulate that there are no duplicate edges or self-loops. Initially, K is set to 4 and p is set to 0.3.

[0203] NW Small World Network: Similar to the K-Nearest Neighbor Network, an edge is added between randomly selected pairs of Agent nodes with probability p. This edge-adding operation also ensures that there are no duplicate edges or self-loops in the network. Initially, K is set to 4 and p is set to 0.3.

[0204] BA scale-free networks: from a network with Starting with a network of nodes, each time a new node is introduced, it interacts with... Connect the existing nodes, that is A new node and an existing node The probability of connection With nodes degree Proportional, that is: After t steps, this algorithm produces a... 1 node A network with edges. Initially, the network is set up with two nodes. Each time a new node is added, it is connected to the two existing nodes in the network according to a priority connection mechanism.

[0205] Step 4: System Output

[0206] The system output section mainly consists of three modules: main interface output, internal network diagram output, and data visualization, as detailed below:

[0207] (1) Main interface output

[0208] During system operation, orders are constantly being transferred between institutions. This involves significant data interaction; after a user clicks "run," the backend needs to continuously send data to the frontend, similar to a phone call where the other party is constantly speaking. Therefore, the main interface uses WebSocket technology for data interaction. When a user clicks "run," the system first checks if the user's browser supports WebSocket. If it does, it initializes a WebSocket. After the frontend's WebSocket successfully establishes a connection with the backend's port 8765, it sends a "begin" message to indicate the start of system operation. If the user's browser does not support WebSocket, a pop-up notification will appear.

[0209] In Echarts' relationship diagrams, order transfers can be represented using links between nodes. For ease of observation by users and testing during development, the same color is used to mark the transfer of the same order, and an animation is added to indicate which organization the order is transferred from and to. The effect is as follows: Figure 5 As shown.

[0210] The system also supports real-time viewing of platform performance, institutional costs and revenues, and the number of order categories arriving in the system during operation, displayed using line charts, bar charts, and column charts respectively. This data arrives along with the order transmission information sent via WebSocket from the backend. Users can view this data by clicking the data display button in the upper right corner. This functionality is implemented using Element-Plus's el-drawer component. To prevent inconsistent data display or delays when opening the drawer multiple times, all data actively pushed from the backend is saved on the frontend.

[0211] (2) Output of internal network diagram of the organization

[0212] The organization's internal output includes a network diagram and some real-time operational data. During system operation, the organization's internal agents continuously process orders arriving at the organization. Data on the main interface is continuously pushed from the backend to the frontend via a WebSocket connection. To save on the overhead of establishing connections between the frontend and backend, internal organization data is also transmitted via the main interface's WebSocket connection and arrives at the frontend simultaneously with the main interface data.

[0213] To visually represent agent states, network diagrams can use different colors to indicate agent idle / busy states, with blue / green representing idle states and striking red representing active agents. For example... Figure 6 As shown, in order to ensure the rigor of the system and meet testing requirements, in some embodiments, users can also point the mouse to the Agent icon, and the system will display the name of the Agent.

[0214] (3) Data Analysis

[0215] The system has an experimental analysis interface. After each round of operation, it outputs the main data of that round, allowing users to more intuitively compare the data results from different rounds.

[0216] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of one or more computer-usable storage media (including, but not limited to, disk storage, etc.) containing computer-usable program code. CD - ROM It takes the form of a computer program product implemented on (such as optical memory, etc.).

[0217] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0218] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0219] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0220] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made by those skilled in the art without creative effort within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A digital service market simulation method, characterized in that, Includes the following steps: Model the external demand environment in the digital service market, map the external demand environment to different types and difficulties of order status representing user needs, and set order arrival strategies and order difficulty generation strategies. The entities of digital service systems in the digital service market are modeled, including platforms, institutions, and executing intelligent agents, forming an entity class table for each entity; Model the order transmission rules between institutions in the digital services market and the collaboration rules of execution agents within institutions; Based on the modeling results of the three parts, the external demand environment, subjects and rule parameters of the target digital service market are configured, and simulation is carried out to obtain simulation results; The system requires real-time data display for order transmission and system performance. WebSocket technology is used for real-time data interaction between the front-end and back-end. For users' one-way access requests to the system, the Flask framework is used for data interaction between the front-end and back-end. Intelligent agents include salespeople and software robots; The process of modeling the order transmission rules between institutions and the collaborative rules of execution agents within institutions in the digital service market includes: the platform is responsible for receiving orders from external sources, allocating them to institutions according to a set algorithm, and the overall efficiency of the platform reflects the quality of the service. Different orders are processed in parallel, while the same order is processed serially. After receiving an order from a platform or other organization, the organization selects an execution agent to process the order according to its internal order allocation strategy. The execution agent has the right to decide whether to establish a connection with other execution agents to jointly process the order. Each executing agent forms a collaborative network. Only executing agents that are connected within the collaborative network can establish connections to jointly process orders. Moreover, only agents with the attribute of "salesperson" can proactively initiate connections to other executing agents.

2. The digital service market simulation method as described in claim 1, characterized in that, The process of mapping the external demand environment to order statuses representing different types and difficulties of user demand includes constructing an order entity attribute table, which includes the order type, name, pending queue, order value to order difficulty ratio, the agent accepting the order, order status, order arrival time, order delivery connection, and upper and lower limits of the order difficulty threshold.

3. The digital service market simulation method as described in claim 1, characterized in that, The process of setting the order difficulty generation strategy includes: the distribution of orders provided by the user includes normal distribution and uniform distribution. In the case of normal distribution, multiple difficulty generation levels are provided to the user based on the length of the order and the overall difficulty, and the order probability of each level is controlled according to the normal distribution point value. In a uniform distribution scenario, users cannot select the difficulty level of the generated order, and the types and difficulty values ​​of generated orders are uniformly distributed.

4. The digital service market simulation method as described in claim 1, characterized in that, The process of modeling the entities of digital service systems in the digital service market includes: The platform is used to coordinate and control the entire digital service system, receive orders from outside the digital service system, allocate orders to different institutions for processing, and calculate total costs and total revenue to evaluate overall efficiency. The organization consists of internal salespersons and software robots, used to receive orders assigned by the platform and to transfer orders between organizations; The executing intelligent agent includes a salesperson and a software robot. The salesperson is configured to handle orders of all difficulty, and the salesperson can create value by completing the order. The value created varies depending on the difficulty, processing time, and quantity of the order. The software robot is configured to handle orders with a difficulty level lower than a set threshold, but its order processing speed is higher than that of a salesperson, and the software robot does not incur additional costs when processing orders.

5. The digital service market simulation method as described in claim 1, characterized in that, The process of allocating orders to institutions according to the set algorithm includes: the platform determines the institutions participating in the allocation based on the type of the order; sorts the institutions participating in the allocation according to the difficulty of the order; and selects the institution with the lowest occupancy rate to prioritize the allocation of the order. If all software robots of the currently allocated institution are already occupied, the order will be prioritized to the institution whose software robots are not fully occupied, according to the allocation order.

6. The digital service market simulation method as described in claim 1, characterized in that, The process of selecting the execution agent to handle the order based on the organization's internal order allocation strategy includes: if the difficulty of the order is less than the difficulty threshold that the software robot can handle, then an idle software robot is used first; if all software robots are occupied, then the execution agent with the fewest unprocessed tasks within the organization is queried to handle the order. If the difficulty of an order exceeds the threshold that the software robot can handle, the system will query the salesperson with the fewest unprocessed tasks within the organization and assign the order to that salesperson for processing.

7. The digital service market simulation method as described in claim 1, characterized in that, After the various intelligent agents form a collaborative network, the rules for processing orders include: If the order difficulty value is less than the difficulty threshold that the software robot can handle, the salesperson will choose other idle execution agents that are connected to it to cooperate in processing; If the difficulty of an order exceeds the threshold that the software robot can handle, the salesperson will choose to collaborate with other available salespersons who are connected to them. If other execution agents that have established connections with this execution agent are all in an occupied state, then this execution agent can only process the order alone.

8. The digital service market simulation method as described in claim 1, characterized in that, When executing an intelligent agent to process an order, the order processing time needs to be calculated based on the difficulty value of the order and the corresponding work efficiency of the intelligent agent. If the difficulty value of the order and the corresponding work efficiency of the intelligent agent are not divisible, the order processing time is calculated by rounding up. If multiple agents collaborate to process an order, the order processing time is calculated based on the order's difficulty level and the sum of the work efficiencies of each participating agent.

9. A digital service market simulation system, employing the digital service market simulation method described in claim 1, characterized in that, include: The external demand environment module is configured to model the external demand environment in the digital service market, map the external demand environment to different types and difficulties of order status representing user needs, and set order arrival strategies and order difficulty generation strategies. The entity modeling module is configured to model the entities of digital service systems in the digital service market, which include platforms, institutions and executive agents, forming an entity class table for each entity; The rule modeling module is configured to model the order transfer rules between institutions in the digital services market and the collaborative rules of execution agents within institutions. The simulation execution module is configured to perform simulations based on the modeling results of three parts: the external demand environment, subjects, and rule parameters of the target digital service market, and to obtain simulation results.

Images