62 results about "Cost evaluation" patented technology

The invention provides a logistics transportation optimization method and system based on a traffic logistics large model, and the method comprises the steps: carrying out the element extraction through receiving a basic scheduling instruction of a logistics transportation task, generating core element information through semantic scene analysis, and carrying out the transportation scene adaption based on the core element information. Generating a scene influence factor set in combination with a depth feature mining result of the historical transportation scene database; then, calling a pre-trained traffic logistics large model to carry out collaborative path decision on the scene influence factor set, and outputting a candidate path scheme population containing a path topological structure and cost feature distribution through interactive iteration of a path generation unit and a cost evaluation unit in the model; and performing multi-round evolution screening on the candidate path scheme population according to a preset optimization target, determining an optimal path scheme meeting transportation requirements, and generating a transportation scheduling instruction set to be pushed to a transportation execution terminal after conversion. According to the invention, the adaptability, decision-making precision and reliability in the logistics transportation optimization process can be effectively improved.

The invention belongs to the technical field of computing resource allocation, and discloses a task-driven computing resource dynamic allocation system, which is characterized in that task differentiation features are extracted through a task feature intelligent analysis module, and global optimal allocation, 'task fingerprint 'matching and opportunity cost evaluation mechanisms are realized in combination with a reinforcement learning scheduling decision module. Mismatching conditions such as allocation of lightweight tasks to a high-video-memory GPU are avoided, meanwhile, the elastic resource pool dynamically adjusts the capacity, reduces resource locking during task pause, remarkably reduces resource waste, improves the overall utilization rate of computing resources such as the GPU, can quickly expand the capacity through the elastic resource pool to cope with burst tasks, and a cross-node resource cooperation module constructs a global view and improves the overall utilization rate of the computing resources. Efficient calling of idle resources is realized; a'logic super node 'and zero-aware migration technology breaks node barriers, avoids global redundancy caused by local optimum, enables resources to flexibly flow along with task loads, and improves the adaptive capacity of a system to dynamic tasks.

The invention discloses an unmanned aerial vehicle trajectory optimization and power distribution method, system and device based on cost evaluation reinforcement learning, and a medium, and the method comprises the steps: constructing an unmanned aerial vehicle data service scene under safety constraints, modeling a multi-objective optimization problem of unmanned aerial vehicle trajectory optimization and power distribution, and designing an unmanned aerial vehicle data service evaluation index; converting a multi-objective optimization problem into a constrained Markov decision problem; respectively designing a state space, an action space, an instant reward function and a corresponding cost function; constructing a cost evaluation critic network parallel to the reward evaluation critic network, training and updating network parameters of the unmanned aerial vehicle by using a cost evaluation SAC algorithm according to a constrained Markov decision problem, and calculating trajectory optimization and power distribution parameters of the unmanned aerial vehicle; the system, the equipment and the medium are used for implementing the method. According to the invention, ground user data service requirements of a data service scene under random and dynamic security constraints are met.

The embodiment of the invention provides an intelligent decision-making method and device for cross-border logistics, electronic equipment and a storage medium, and is applied to the technical field of logistics decision-making. The intelligent decision-making method for cross-border logistics comprises the steps of collecting heterogeneous logistics data according to user demand information, performing language entity alignment processing on the heterogeneous logistics data to obtain standardized shipping data, performing transportation cost evaluation according to the standardized shipping data to generate comprehensive cost data, performing path search according to the comprehensive cost data through a heuristic algorithm to obtain an initial feasible path set, and performing multi-target optimization screening processing on the initial feasible path set to generate an initial logistics planning scheme set; and carrying out risk simulation and TOPSIS preference sorting on the initial logistics planning scheme set to obtain an optimal logistics planning scheme and a corresponding reported value. According to the invention, through cost evaluation, constraint path search, multi-target optimization and risk quantification, a more accurate transportation scheme is provided for the user, and the cross-border logistics operation efficiency is improved.

The invention discloses a GPU cluster resource allocation method and device, equipment and a storage medium, and relates to the technical field of GPU cluster resource allocation. According to the method, a hardware performance description vector is generated by fusing static hardware parameters and dynamic micro-benchmark test data, the limitation that the GPU performance is represented only by depending on the static parameters is broken through, and the actual operation capacity of different architecture GPUs in a heterogeneous cluster is accurately matched; task feature vectors are generated by extracting task calculation features and resource demands, and quantitative description of reasoning task resource consumption features is achieved; information of hardware, tasks and load dimensions is integrated through a machine learning model, performance degradation characteristics during multi-task parallel are effectively captured, and the accuracy of execution time prediction is improved; scheduling decision operation including candidate node screening, comprehensive cost evaluation and resource reservation backfilling is executed in combination with the predicted execution time, and the task execution efficiency and cluster load balancing are both considered.

When a trigger event causing the priority change of the production task is recognized, the influence degree of the trigger event on an overall production plan is evaluated, whether the influence degree of the trigger event on the overall production plan is low or not is judged, and when the judgment result is no, the priority of the production task is changed according to the priority of the production task. Generating a plurality of candidate priority change schemes, performing priority change cost evaluation on the candidate priority change schemes, and determining a target priority change scheme in the candidate priority change schemes according to a priority change cost evaluation result, the priority of the production task is adjusted according to the target priority change scheme, the optimal production priority change scheme can be produced in combination with the actual situation, and the adverse effect caused by production priority change is effectively reduced.

The invention discloses a data-intensive workflow dynamic task allocation method and system oriented to cloud-edge cooperative computing, and relates to the technical field of cloud computing and edge computing collaboration.The method comprises the steps that workflow and cloud-edge resources are modeled in a unified mode, and a comprehensive cost function containing time, economy and energy consumption is defined; analyzing the workflow to obtain a conflict task set and a maximum data path set; in combination with resources and network states monitored in real time, candidate nodes are filtered according to data privacy constraints, edges or cloud nodes with the minimum comprehensive cost are selected for tasks for allocation, and sorting is performed according to priorities during multiple tasks; the system comprises a resource monitoring module, a workflow analysis and preprocessing module, a cost evaluation module and a dynamic scheduler module, the modules cooperate to realize dynamic allocation, the problems of resource heterogeneity, network dynamic and the like under cloud edge cooperation are solved, multi-objective optimization and privacy protection are balanced, and the workflow execution efficiency is improved.

The invention discloses an intelligent construction management system and method based on data analysis, and belongs to the technical field of construction management, and the method comprises the steps: obtaining original plan feature data and change estimation feature data of a target change task in a construction project, and generating a basic change cost evaluation index of the target change task; establishing a task relationship network among all tasks in the construction project; obtaining the number of affected tasks actually affected by the target change task in the completed tasks of the construction project, and generating a target change task change feasibility evaluation index in combination with the target change task basic change cost evaluation index; obtaining a change reason of the target change task in the construction project, and generating a final change feasibility evaluation index of the target change task in combination with the change feasibility evaluation index of the target change task, thereby managing the target change task in the construction project; according to the invention, construction resource distribution and progress control precision can be optimized.

The embodiment of the application provides a neural network model configuration determination method and device, equipment and a storage medium, according to the dimension information of the target network layer and the hardware instruction of the target hardware, the prior information and the universal information in the preset model cost evaluation table are inquired, the preset configuration that makes the number of clock cycles of each target network layer shortest can be obtained, that is, the configuration of each target network layer when the number of clock cycles of the target neural network model is shortest, so that the optimal configuration of the target neural network model in the target hardware is obtained, and the optimal configuration can be used to realize the rapid landing of the neural network model in the target hardware subsequently.

The invention provides a modular design and production method for assembly type ice and snow building components based on cost and manufacturability analysis. According to the method, the cost-manufacturability evaluation curve is automatically generated in the dividing and checking process, the optimal cost solution and the machinable boundary are output in a visual mode, and a basis is provided for design decision making of the ice shell structure. A data file output by the method comprises discrete unit geometric parameters, mold numbers, machining sizes, cost evaluation results and assembly sequences, and can be directly used for follow-up mechanical arm path planning and mold production.