Research on Optimization Scheduling Algorithm of Engineering Project Resources Based on Cloud Computing

Aiming at the problem of resource optimization scheduling for large and complex engineering projects, this paper proposes an optimization scheduling algorithm system based on cloud computing, which deeply integrates cloud computing, convolutional neural network (CNN) and Beetle Antennae Search (BAS). The Cloud Convolution Antennae Search (CCAS) model is proposed to realize efficient, accurate and dynamic resource allocation and scheduling. With cloud computing as the underlying support, the CCAS model realizes efficient storage, processing and real-time update of large-scale engineering project data through the powerful computing power and elastic expansion characteristics of cloud computing, ensuring that the model can still maintain efficient operation in the face of massive and dynamic data, and providing a solid foundation for subsequent intelligent decision-making. Secondly, at the algorithmic level, CNN is applied to the prediction of resource demand of engineering projects. It uses its advantages in spatiotemporal series data modeling to carry out deep learning on multi-dimensional data such as project schedule and resource consumption to achieve accurate prediction of future resource demand. At the same time, a Longniu whisker search algorithm is introduced to make resource scheduling decision. BAS simulates the natural process of Longurus searching for food, explores the scheduling space effectively by iteratively updating the search path, and finds the global optimal or near-optimal scheduling scheme. The organic combination of the three not only provides strong computing power support and the advantages of data management and deep learning in complex data mining, but also utilizes the efficiency of optimization algorithms in large-scale problem solving. In terms of system architecture, the three-layer cloud service model based on IaaS, PaaS and SaaS is built to provide comprehensive support for the research and development, deployment, operation and application of resource scheduling algorithms. IaaS layer integrates hardware resources to realize on-demand infrastructure rental. PaaS layer provides big data processing, AI algorithm library and other development environment to help scheduling model construction and optimization; The SaaS layer encapsulates various scheduling functions and is easy for engineering project teams to call directly. Finally, this paper compares the model algorithm with the neural network and analyzes the case to verify the significant advantages of the proposed system in the accuracy of project resource demand prediction, the scientific scheduling decision and the overall efficiency improvement. The results show that the cloud computing based engineering project resource optimization scheduling algorithm system integrates the CCAS of the three, effectively solves the complexity and uncertainty problems that the traditional scheduling methods are difficult to deal with, realizes the intelligent and fine management of engineering projects, and provides a strong technical support for industry practice.