Project management resource optimization configuration platform

Abstract

The invention relates to the technical field of engineering management, and discloses an engineering management resource optimization configuration platform, which comprises the following modules: a data acquisition and management module used for acquiring real-time state data of manpower, equipment and materials in an engineering project through Internet of Things equipment and sensors; the resource pooling and sharing module is used for carrying out centralized management on resources in the engineering projects, constructing a cross-project resource pool and carrying out resource sharing according to project requirements and resource states; and the intelligent scheduling and optimizing module is used for dynamically scheduling resources based on project requirements, task priorities and task dependency relationships. By adopting an intelligent scheduling method combining a hybrid genetic algorithm and a particle swarm optimization algorithm and a deep reinforcement learning dynamic scheduling module, resources can be dynamically allocated according to project requirements and task priorities, the utilization rate of the resources can be maximized, the phenomenon that the resources are idle or reused is reduced, and efficient configuration of the resources is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering management, in particular to an engineering management resource optimization configuration platform. Background Art

[0002] In today's large-scale engineering project management, as the scale and complexity of projects increase, how to efficiently manage and allocate resources has become the key to the success of the project. There are many types of resources in engineering projects, including human resources, equipment resources, and material resources. The reasonable allocation of these resources directly affects the progress, quality, and cost of the project. Therefore, an intelligent engineering management resource optimization and allocation platform came into being. The goal of this platform is to use advanced technical means to optimally allocate and intelligently schedule the resources of multiple projects in a dynamic environment, so as to improve the overall efficiency of the project, reduce costs, reduce resource waste, and ensure that the project is delivered on time.

[0003] Traditional resource allocation methods for engineering management mostly rely on manual experience for resource allocation and scheduling. This model shows great limitations when facing the modern complex multi-project parallel environment. Usually, project managers can only make decisions based on experience and local information, resulting in low resource utilization and inability to achieve cross-project resource sharing and dynamic adjustment. Especially in the case of tight resources, it is easy for some projects to be short of resources while other projects have excess resources. In addition, traditional resource management platforms lack real-time data collection and analysis capabilities and cannot quickly reflect changes in the project, further increasing the difficulty of resource scheduling.

[0004] To this end, those skilled in the art provide an engineering management resource optimization configuration platform to solve the problems raised in the above background technology. Summary of the invention

[0005] In view of the deficiencies of the prior art, the present invention provides an engineering management resource optimization configuration platform, which solves the problems of low configuration resource utilization and slow response speed in the prior art.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an engineering management resource optimization configuration platform, including the following modules:

[0007] Data collection and management module, which is used to collect real-time status data of manpower, equipment and materials in engineering projects through IoT devices and sensors;

[0008] Resource pooling and sharing module, used to centrally manage resources in the engineering project, build cross-project resource pools, and share resources according to the needs and resource status of each project;

[0009] Intelligent scheduling and optimization module, which is used to dynamically schedule resources based on project requirements, task priorities and task dependencies, and generate the optimal resource allocation plan;

[0010] Deep reinforcement learning dynamic scheduling module, which is used to dynamically optimize resource scheduling based on the Markov decision process model, learn and apply the best scheduling strategy;

[0011] The risk management and early warning module is used to evaluate the risks of resource shortages and task delays in projects in real time, and generate early warning information when risks are detected.

[0012] Preferably, the data collection and management module uses Internet of Things sensors, including RFID devices, GPS modules and equipment status sensors, for real-time collection of personnel location, equipment status and material inventory data.

[0013] Preferably, the data collection and management module includes:

[0014] The real-time monitoring and data verification submodule is used to monitor and verify the collected real-time data to ensure the accuracy and completeness of the data. When the data is abnormal, an alarm is automatically generated and the collection strategy is adjusted;

[0015] The data storage and historical data analysis submodule uses big data analysis technology to conduct deep learning and trend prediction on the collected historical data to optimize future resource scheduling decisions.

[0016] Preferably, the resource pooling and sharing module classifies various types of resources into human resource pools, equipment resource pools and material resource pools through a resource pooling model, and dynamically allocates and prioritizes resources according to their type, availability, location and cost.

[0017] Preferably, the resource pooling and sharing module includes:

[0018] The resource conflict detection and resolution submodule is used to detect conflicting demands for the same resource in multiple projects, and adjust resource allocation plans based on resource priority, task urgency and task dependency to ensure efficient use of resources.

[0019] Preferably, the intelligent scheduling and optimization module adopts a combination of a hybrid genetic algorithm and a particle swarm optimization algorithm to optimize resource allocation schemes for multiple projects through global search and local optimization.

[0020] Preferably, the fitness function of the hybrid genetic algorithm is defined according to the scheduling cost, task completion time and resource utilization of the project to minimize the scheduling cost and delay time and maximize the resource utilization. The fitness function f(x) is determined by the following formula:

[0021]

[0022] Where: c ij For resource R j Assigned to Project P i The scheduling cost of i ) is the project P i The task completion time U(R j ) is the resource R j ; α, β and γ are weight factors that control scheduling cost, task completion time and resource utilization respectively.

[0023] Preferably, the deep reinforcement learning dynamic scheduling module performs resource scheduling based on a deep Q network, adopts Markov decision process modeling, and constructs a scheduling strategy with state space, action space and reward function, and the reward function is:

[0024]

[0025] Where: U(R) is the resource utilization rate; C(R,P) is the scheduling cost of resource R in project P; T(P) is the task completion time of project P, S(a) is the task progress speed brought by the current scheduling strategy; λ1 and λ2 are the weight factors of resource utilization and task progress speed respectively.

[0026] Preferably, the deep reinforcement learning dynamic scheduling module includes:

[0027] The experience replay mechanism submodule is used to save past scheduling decisions and result data during the resource scheduling process, and optimize the scheduling strategy based on the experience replay mechanism.

[0028] A device for optimizing the configuration of engineering management resources is provided, which is equipped with a platform for optimizing the configuration of engineering management resources. The device comprises a processor, a memory, a network interface and an input-output interface.

[0029] The present invention provides a project management resource optimization configuration platform, which has the following beneficial effects:

[0030] 1. The present invention adopts an intelligent scheduling method that combines a hybrid genetic algorithm with a particle swarm optimization algorithm, as well as a deep reinforcement learning dynamic scheduling module, which can not only dynamically allocate resources according to project requirements and task priorities, but also maximize resource utilization, reduce resource idleness or reuse, and ensure efficient resource allocation.

[0031] 2. The present invention can effectively predict and respond to the risks of resource shortages and task delays through real-time risk management and early warning modules. The system can make scheduling adjustments and optimize resource allocation in advance, thereby avoiding cost increases and project progress delays caused by insufficient resources or delays, and ensuring that the project is completed on time and within budget.

[0032] 3. The present invention optimizes the scheduling strategy based on the Markov decision process and the deep Q network. The present invention can continuously learn and optimize resource scheduling decisions. The system continuously improves its adaptability to resource scheduling in complex environments through historical data and experience playback mechanism, thereby providing more intelligent, flexible and efficient decision support for the parallel management of multiple projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the configuration platform framework flow of the present invention;

[0034] Figure 2 This is a schematic diagram of the data acquisition and management module framework of the present invention;

[0035] Figure 3 It is a schematic diagram of the resource pooling and sharing module framework of the present invention;

[0036] Figure 4 Schematic diagram of the deep reinforcement learning dynamic scheduling module framework of the present invention. DETAILED DESCRIPTION

[0037] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] Embodiment 1:

[0039] Please refer to the attached Figure 1 The embodiment of the present invention provides an engineering management resource optimization configuration platform, including the following modules:

[0040] Please refer to the attached Figure 2 , the data collection and management module is used to collect real-time status data of manpower, equipment and materials in the project through IoT devices and sensors; the data collection and management module uses IoT sensors, including RFID devices, GPS modules and equipment status sensors, to collect real-time personnel location, equipment status and material inventory data;

[0041] The data collection and management modules include:

[0042] The real-time monitoring and data verification submodule is used to monitor and verify the collected real-time data to ensure the accuracy and completeness of the data. When the data is abnormal, an alarm is automatically generated and the collection strategy is adjusted;

[0043] The data storage and historical data analysis submodule uses big data analysis technology to conduct deep learning and trend prediction on the collected historical data to optimize future resource scheduling decisions.

[0044] Specifically, the data collection and management module plays a vital role in the engineering management resource optimization configuration platform. It is responsible for real-time collection and monitoring of the status of various resources in the project through IoT devices and sensors. By integrating RFID devices, GPS modules and equipment status sensors, this module can fully capture and transmit key data related to project execution. For example, RFID devices can accurately record the quantity and location of materials in the material entry and exit process, ensuring that the material inventory is updated in real time, which is important for avoiding material shortages or waste. The GPS module is used to track the geographic location of on-site construction personnel and the movement trajectory of equipment, so that managers can grasp the spatial distribution and dynamic changes of various resources, so as to respond quickly when resource demand conflicts occur. The equipment status sensor collects the operating status and performance data of the construction equipment in real time, identifies possible equipment failures in advance, and performs maintenance in a timely manner, thereby avoiding construction delays caused by equipment failures.

[0045] This module further integrates the real-time monitoring and data verification submodules to ensure the accuracy and real-time nature of the data. In actual operations, engineering management often relies on multiple sensor devices working simultaneously. Therefore, data accuracy is the key to ensuring the rational allocation and scheduling of resources. This submodule works synchronously with multiple IoT sensors to monitor the data transmitted by the sensors in real time and automatically detect abnormal data. For example, when the equipment sensor feedback abnormal operating status, or when the material inventory data fluctuates abnormally, the system will immediately generate an alarm to prompt the management personnel to take intervention measures, or automatically adjust the working status of the collection equipment according to the preset response strategy. This automatic verification mechanism greatly improves the reliability of the data, especially in complex engineering environments, and avoids data deviations caused by human errors or sensor failures.

[0046] In addition, the data storage and historical data analysis submodules included in this module are not only responsible for the long-term storage of data, but also use big data technology to conduct in-depth analysis of past resource usage data. The engineering management resource optimization configuration platform can generate trend forecasts and assist future decision-making by analyzing data such as equipment operation status, task completion time, and manpower scheduling in historical projects. For example, the system can predict the possible peak manpower demand in future projects based on the use of human resources in multiple previous projects, and reserve and schedule human resources in advance. This process can effectively avoid cost waste caused by shortage or overuse of human resources. At the same time, historical data analysis can also help identify common equipment failure modes and their causes in different types of projects. The system will automatically update the scheduling strategy to ensure that the equipment is effectively maintained in the appropriate time period, extend the service life of the equipment, and avoid the impact of equipment failure on the construction period.

[0047] In actual operation, the real-time monitoring and data verification submodule works closely with the data storage and historical data analysis submodule to ensure the stability of the platform operation and the intelligent scheduling effect. The real-time monitoring submodule first ensures that all collected data is accurate and correct, and can quickly respond to emergencies in the on-site environment, such as personnel movement, equipment abnormalities, or material shortages; while the historical data analysis submodule generates trend forecasts and resource scheduling optimization plans by analyzing a large amount of historical data, and provides valuable experience accumulation for the execution of future projects. The combination of the two enables the platform to not only adapt to the real-time changing construction environment, but also has the ability to plan and optimize resources from a long-term perspective.

[0048] Through this synergy, the data collection and management module not only provides reliable basic data support for other modules, but also lays the foundation for the intelligent and optimized resource scheduling. The entire system can achieve efficient resource allocation and scheduling in the actual engineering management process, improve the efficiency of overall project management, reduce operating costs, and ensure that the project is completed with high quality within the specified time. This dynamic connection, data flow and information sharing between modules enable the engineering management resource optimization configuration platform to provide accurate resource optimization solutions based on actual engineering needs, ensuring that all types of resources are optimally scheduled and used.

[0049] Please refer to the attached Figure 3 , the resource pooling and sharing module is used to centrally manage resources in engineering projects, build cross-project resource pools, and share resources according to the needs of each project and resource status; the resource pooling and sharing module classifies various types of resources into human resource pools, equipment resource pools, and material resource pools through a resource pooling model, and dynamically allocates and prioritizes resources based on their type, availability, location, and cost.

[0050] Resource pooling and sharing modules include:

[0051] The resource conflict detection and resolution submodule is used to detect conflicting demands for the same resource in multiple projects, and adjust resource allocation plans based on resource priority, task urgency and task dependency to ensure efficient use of resources.

[0052] Specifically, the resource pooling and sharing module in the engineering management resource optimization and configuration platform is mainly used to centrally manage the resources of different projects, and to achieve efficient sharing and dynamic allocation of resources by building a cross-project resource pool. This module classifies and manages various types of resources such as manpower, equipment and materials in the project through a resource pooling model, and builds a human resource pool, an equipment resource pool and a material resource pool respectively to ensure that the utilization efficiency of resources is maximized and provide a shareable resource reserve for multiple projects. In actual operation, the resource pooling and sharing module conducts a comprehensive evaluation based on multiple dimensions such as resource type, availability, geographical location and cost, and dynamically allocates resources to ensure that resources can be used efficiently and reasonably according to the actual needs of each project. For example, a certain equipment can be quickly allocated to project B after completing the task in project A, thus avoiding the problem of idle resources or repeated procurement. Through this centralized management method, the platform can achieve unified resource scheduling across projects, optimize resource allocation efficiency, and save overall project costs.

[0053] The core function of the resource pooling and sharing module is to ensure the effective use of resources by prioritizing and dynamically allocating resources. Specifically, the platform will prioritize projects based on factors such as the urgency of the project, task dependencies, and resource availability. For example, for critical tasks or projects approaching deadlines, the system will prioritize the required resources; for non-critical tasks or slow-moving projects, the system can appropriately postpone resource allocation. This priority-based allocation mechanism ensures that resources are most reasonably scheduled throughout the project cycle, avoiding resource waste or unnecessary delays.

[0054] In order to further improve the efficiency of resource utilization, the resource pooling and sharing module also includes a resource conflict detection and resolution submodule. The main function of this submodule is to detect resource allocation conflicts when multiple projects have demands for the same resource at the same time, and automatically adjust according to the preset scheduling rules. In actual applications, different engineering projects may have competing demands for the same equipment or materials because of the same resource type or close location. For example, multiple projects may need to use a specific equipment for construction at the same time. At this time, the conflict detection submodule will automatically identify this resource conflict and adjust the resource allocation strategy according to the urgency of the task, dependencies, and the progress of other projects. The system will give priority to allocating resources to projects with urgent tasks and strong dependencies, while other projects may be postponed or allocated alternative resources.

[0055] This conflict resolution mechanism can not only detect resource conflicts through real-time data, but also conduct a comprehensive evaluation based on the task dependencies set within the platform. If the completion of a task directly depends on the progress of another project or task, the system will prioritize the smooth completion of the task of the dependency, thereby avoiding the entire project progress being blocked due to delays in dependent tasks. In addition, the platform can also dynamically adjust the order of resource allocation by weighing the cost of resources. For example, when the cost of a certain resource is high, the system can arrange the efficient use of the resource among multiple projects through a resource sharing mechanism, thereby reducing the total cost.

[0056] Through the resource conflict detection and resolution submodule, the platform can automatically generate solutions and make real-time adjustments when resource demand conflicts occur, ensuring that project progress is not hindered. The automated adjustment function of this module greatly reduces delays or misjudgments caused by manual scheduling and improves the response efficiency in engineering management. For example, the platform can quickly identify and provide optimization solutions when resource conflicts occur, recommend the use of alternative equipment or materials, or delay the scheduling of resource use for certain non-urgent projects, thereby freeing up resources for critical tasks. This series of dynamic adjustments ensures the maximum utilization of resources and improves the coordination ability when multiple projects are carried out in parallel.

[0057] Please refer to the attached Figure 1 , the intelligent scheduling and optimization module is used to dynamically schedule resources based on project requirements, task priorities and task dependencies, and generate the optimal resource allocation plan; the intelligent scheduling and optimization module adopts a combination of hybrid genetic algorithm and particle swarm optimization algorithm to optimize the resource allocation plans of multiple projects through global search and local optimization.

[0058] The fitness function of the hybrid genetic algorithm is defined according to the project scheduling cost, task completion time and resource utilization to minimize the scheduling cost, delay time and maximize resource utilization. The fitness function f(x) is determined by the following formula:

[0059]

[0060] Where: c ij For resource R j Assigned to Project P i The scheduling cost of i ) is the project P i The task completion time U(R j ) is the resource R j ; α, β and γ are weight factors that control scheduling cost, task completion time and resource utilization respectively.

[0061] Specifically, the intelligent scheduling and optimization module uses a hybrid genetic algorithm and particle swarm optimization algorithm (PSO) to organically combine global search and local optimization. The global search capability of the genetic algorithm can find near-optimal solutions in a vast solution space by simulating natural selection and evolutionary processes. The particle swarm optimization algorithm quickly converges to the local optimal solution by simulating the collaborative behavior of biological groups. The combination of these two algorithms ensures that the intelligent scheduling module can find the global optimal solution and maintain high efficiency and rapid response capabilities in the local optimization process when facing complex and changing project scheduling requirements. Especially when facing scenarios where multiple projects compete for resources at the same time, the hybrid algorithm can ensure the accuracy of scheduling decisions and avoid waste of resources or unnecessary scheduling conflicts.

[0062] In the process of intelligent scheduling and optimization, the fitness function of the hybrid genetic algorithm is the key to determining the optimization effect of the algorithm. The goal of the fitness function is to evaluate the scheduling cost, task completion time and resource utilization of the project, and generate the optimal resource allocation plan by weighing these factors.

[0063] In practical applications, scheduling costs are the direct economic costs incurred in the allocation process of resources, such as equipment rental costs, material consumption costs, or labor costs. This factor is crucial in engineering management because the increase in scheduling costs will directly affect the total budget of the project. By setting appropriate weight factors, the system can balance cost control and resource utilization efficiency.

[0064] Task completion time is the time required to complete tasks in the project's critical path, which reflects the urgency of the project progress.

[0065] The intelligent scheduling module will reasonably allocate resources to shorten the delay time of key tasks based on the priority of tasks and the duration constraints of the project. In a project, the delay of certain tasks may have a chain reaction on the entire project, so the fitness function controls the weight of the task completion time through parameters to ensure that the project can be completed on time and avoid project losses caused by task delays.

[0066] Resource utilization reflects the efficiency of the allocated resources in performing tasks. The system ensures that all equipment, manpower or materials are fully utilized during the scheduling process by improving resource utilization, thereby reducing the idle time of resources or the problem of excessive reuse rate. By controlling the weight of resource utilization by parameters, the system can further optimize the scheduling efficiency of resources while meeting project requirements. In order to optimize actual operations, the intelligent scheduling and optimization module not only evaluates the scheduling plan through the above fitness function, but also uses the local optimization capability of the particle swarm optimization algorithm to dynamically adjust the use of resources in a short period of time. In each scheduling cycle, the system will calculate the optimal resource reconfiguration plan based on the current resource allocation status and task priority to ensure efficient use of resources. This module can adapt to real-time changes in project requirements, give priority to key tasks when resources are tight, and reasonably allocate resources to improve overall and utilization when resources are sufficient.

[0067] Please refer to the attached Figure 4 ,Deep reinforcement learning dynamic scheduling module, which is used to dynamically optimize resource scheduling based on the Markov decision process model, learn and apply the best scheduling strategy;

[0068] The deep reinforcement learning dynamic scheduling module performs resource scheduling based on the deep Q network, adopts Markov decision process modeling, and constructs the scheduling strategy based on the state space, action space and reward function. The reward function is:

[0069]

[0070] Where: U(R) is the resource utilization rate; C(R,P) is the scheduling cost of resource R in project P; T(P) is the task completion time of project P, S(a) is the task progress speed brought by the current scheduling strategy; λ1 and λ2 are the weight factors of resource utilization and task progress speed respectively.

[0071] The deep reinforcement learning dynamic scheduling module includes:

[0072] The experience replay mechanism submodule is used to save past scheduling decisions and result data during the resource scheduling process, and optimize the scheduling strategy based on the experience replay mechanism.

[0073] Specifically, the deep reinforcement learning dynamic scheduling module is one of the core modules in the engineering management resource optimization configuration platform. It is responsible for intelligent and real-time optimization and scheduling of resources based on complex project requirements and dynamic environments. This module uses the Markov decision process model (MDP) to model the resource scheduling problem as a series of interactive processes of states, actions, and rewards, and continuously optimizes the scheduling strategy through reinforcement learning methods. In the specific implementation, the deep reinforcement learning dynamic scheduling module uses the **deep Q network (DQN) to perform resource scheduling, that is, using a deep neural network to estimate the Q value function, thereby continuously improving the scheduling strategy.

[0074] In this process, the optimization of resource scheduling is based on the specific task status of the project, resource availability, and changes in project requirements. By defining the state space of the system, DQN can identify the state of the system at each scheduling moment, make scheduling decisions based on the current state, and select appropriate actions to adjust resource allocation. This series of actions forms an action space, which is the set of resource scheduling solutions that the system can execute under a given state. After each resource scheduling, the system evaluates the pros and cons of the scheduling strategy based on the actual execution results, calculates the corresponding reward function, and provides feedback on the scheduling effect and updates the strategy.

[0075] In deep reinforcement learning, the reward function is the key evaluation criterion for resource scheduling. The reward function should comprehensively consider resource utilization and task completion speed to score the scheduling strategy. In the process of resource allocation, the system gradually learns which strategies can maximize the progress of the project while ensuring efficient resource utilization by constantly trying different scheduling schemes. For example, the resource utilization rate U(R) refers to whether the resources are fully utilized. The system will give priority to those schemes that can maximize the efficiency of resource utilization; while the task completion time T(P) reflects the impact of the scheduling strategy on the project progress. The system will reward those scheduling schemes that can complete the task in a shorter time. By controlling the proportion of λ1 and λ2, the system can flexibly adjust the balance between resource utilization and task progress. In actual operation, the deep reinforcement learning dynamic scheduling module not only relies on immediate status and feedback to adjust the strategy, but also includes a very important submodule-the experience playback mechanism submodule. The role of this submodule is to save historical scheduling decisions and corresponding result data and optimize based on these experiences. In the process of resource scheduling, the platform will record the status, action and result of each scheduling and store this information in the experience library. When the system updates its strategy, the experience replay mechanism allows the system to randomly extract historical data from the experience library for replay, and learn and optimize these data. The advantage of this mechanism is that it can avoid the problem of unstable strategy updates caused by reinforcement learning and improve learning efficiency. The experience replay mechanism reviews past decisions and feedback, allowing the deep Q network to extract rules from decision information at different time points, thereby more effectively optimizing scheduling strategies. Especially when facing multiple projects and complex scheduling requirements, the experience replay mechanism can help the system better deal with resource conflicts and resource competition issues. For example, when a scheduling strategy in historical records performs poorly in similar resource competition scenarios, the system can avoid repeated errors by replaying the experience and optimize resource allocation decisions in future scheduling. The introduction of this submodule enables the system to not only learn the optimal strategy in the current environment in each decision, but also continuously optimize resource scheduling capabilities in a wider range of historical data.

[0076] Through the deep reinforcement learning dynamic scheduling module, the system can optimize resource scheduling in complex environments, and the scheduling strategy will become more and more accurate and efficient over time. In complex engineering projects, this continuous learning ability can significantly improve resource utilization efficiency, reduce scheduling costs, and ensure that projects are completed on time. The experience playback mechanism in the platform ensures that the system can accelerate the learning process through past experience and avoid repeated mistakes on similar problems, thereby further improving the intelligence level and response speed of scheduling.

[0077] Please refer to the attached Figure 1,The risk management and early warning module is used to assess the ,risks of resource shortage and task delay in the project in real time, and generate ,earning information when risks are detected.

[0078] Specifically, in the implementation, the risk management and early warning module first obtains the latest progress data and resource usage of the project in real time through close integration with other modules of the platform. For example, the module can obtain the working status of the current equipment, material inventory and distribution data of human resources from the data acquisition and management module; it can also obtain the resource bottleneck information that may appear in the scheduling process from the intelligent scheduling and optimization module. Through the comprehensive analysis of these data, the system can establish a real-time risk assessment model to predict the deviation of task progress and abnormal resource usage.

[0079] Once the risk of potential resource shortage or task delay is detected, the system will generate early warning information based on the severity and scope of the risk. These early warning information may include: resource shortage warning, task delay warning, equipment failure risk warning, etc. The system classifies these early warning information into three levels: low, medium, and high according to the preset risk level. For high-risk warnings, the system can automatically send notifications to project managers, or automatically adjust resource scheduling according to preset emergency strategies. For example, when it is detected that a key equipment is about to fail, or a key material inventory is about to run out, the system can automatically trigger the corresponding maintenance or replenishment tasks to ensure that the project will not stagnate due to the sudden lack of resources.

[0080] In the process of assessing resource shortages, the risk management and early warning module uses the data analysis function in the platform and combines historical data to predict trends and identify various potential factors that may lead to resource shortages. For example, the system can analyze the resource usage patterns in similar situations in historical projects, predict the resource consumption peak that may occur in the current project, and generate early warning information in advance. If the consumption rate of a certain type of resource exceeds expectations, the system can suggest that project managers increase the supply of the resource, or coordinate with other projects in advance for resource sharing to avoid sudden resource shortages affecting project progress.

[0081] Similarly, task delay risk assessment is also an important function of the risk management and early warning module. During the execution of a project, there are often dependencies between tasks. The delay of a task may affect the progress of subsequent tasks, thereby affecting the delivery time of the entire project. This module continuously monitors the progress of project tasks and combines the task priority and dependency data provided by the intelligent scheduling and optimization module to calculate the completion time deviation of each task in real time and predict the overall progress risk of the project in advance based on the risk assessment model. If it is detected that the progress of a task is significantly behind schedule, the system will immediately generate a warning message for task delays and recommend corresponding countermeasures, such as allocating more resources to speed up the progress of the task, or adjusting the schedule of other tasks to reduce the impact of delays.

[0082] In addition, the risk management and early warning module can also be combined with the deep reinforcement learning dynamic scheduling module to use the learning model to further improve the accuracy of risk prediction. Through the experience data accumulated in historical projects, the system can gradually learn the best response strategies under different risk scenarios, making future risk management more accurate and intelligent. For example, based on the risk handling experience of past projects, the system can give priority to proven response measures when similar risks occur, thereby shortening the response time and reducing the probability of risk occurrence.

[0083] Embodiment 2:

[0084] Based on the same inventive concept, the present invention proposes an engineering management resource optimization configuration device, and the terminal includes: the device includes a processor, a memory, a network interface and an input and output interface; the processor is used to execute instructions in the memory and execute the image recognition method of the above embodiment.

[0085] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0086] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A project management resource optimization configuration platform, characterized in that: Includes the following modules: Data collection and management module, which is used to collect real-time status data of manpower, equipment and materials in engineering projects through IoT devices and sensors; Resource pooling and sharing module, used to centrally manage resources in the engineering project, build cross-project resource pools, and share resources according to the needs and resource status of each project; Intelligent scheduling and optimization module, which is used to dynamically schedule resources based on project requirements, task priorities and task dependencies, and generate the optimal resource allocation plan; Deep reinforcement learning dynamic scheduling module, which is used to dynamically optimize resource scheduling based on the Markov decision process model, learn and apply the best scheduling strategy; The risk management and early warning module is used to evaluate the risks of resource shortages and task delays in projects in real time, and generate early warning information when risks are detected.

2. The engineering management resource optimization configuration platform according to claim 1, characterized in that: The data collection and management module uses IoT sensors, including RFID devices, GPS modules and equipment status sensors, to collect personnel location, equipment status and material inventory data in real time.

3. The engineering management resource optimization configuration platform according to claim 2, characterized in that: The data acquisition and management module includes: The real-time monitoring and data verification submodule is used to monitor and verify the collected real-time data to ensure the accuracy and completeness of the data. When the data is abnormal, an alarm is automatically generated and the collection strategy is adjusted; The data storage and historical data analysis submodule uses big data analysis technology to conduct deep learning and trend prediction on the collected historical data to optimize future resource scheduling decisions.

4. The engineering management resource optimization configuration platform according to claim 1, characterized in that: The resource pooling and sharing module classifies various types of resources into human resource pools, equipment resource pools and material resource pools through a resource pooling model, and dynamically allocates and prioritizes resources according to their type, availability, location and cost.

5. The engineering management resource optimization configuration platform according to claim 1, characterized in that: The resource pooling and sharing module includes: The resource conflict detection and resolution submodule is used to detect conflicting demands for the same resource in multiple projects, and adjust resource allocation plans based on resource priority, task urgency and task dependency to ensure efficient use of resources.

6. The engineering management resource optimization configuration platform according to claim 1, characterized in that: The intelligent scheduling and optimization module adopts a hybrid genetic algorithm and a particle swarm optimization algorithm to optimize the resource allocation schemes of multiple projects through global search and local optimization.

7. The engineering management resource optimization configuration platform according to claim 6, characterized in that: The fitness function of the hybrid genetic algorithm is defined according to the scheduling cost, task completion time and resource utilization of the project to minimize the scheduling cost, delay time and maximize resource utilization. The fitness function f(x) is determined by the following formula: Where: c ij For resource R j Assigned to Project P i The scheduling cost of i ) is the project P i The task completion time U(R j ) is the resource R j ; α, β and γ are weight factors that control scheduling cost, task completion time and resource utilization respectively.

8. The engineering management resource optimization configuration platform according to claim 1, characterized in that: The deep reinforcement learning dynamic scheduling module performs resource scheduling based on the deep Q network, adopts Markov decision process modeling, and constructs a scheduling strategy based on the state space, action space and reward function. The reward function is: Where: U(R) is the resource utilization rate; C(R, P) is the scheduling cost of resource R in project P; T(P) is the task completion time of project P, S(a) is the task progress speed brought by the current scheduling strategy; λ1 and λ2 are the weight factors of resource utilization and task progress speed respectively.

9. The engineering management resource optimization configuration platform according to claim 8, characterized in that: The deep reinforcement learning dynamic scheduling module includes: The experience replay mechanism submodule is used to save past scheduling decisions and result data during the resource scheduling process, and optimize the scheduling strategy based on the experience replay mechanism.

10. A device for optimizing the configuration of engineering management resources, equipped with a platform for optimizing the configuration of engineering management resources according to any one of claims 1 to 9, characterized in that: The device includes a processor, a memory, a network interface, and an input and output interface.

Images