Research on Optimization Model and Algorithm of Major Project Construction Management Based on Unmanned Aerial Vehicles
With the development of unmanned aerial vehicles(UAVs)technology and the reduction of manufacturing cost,the advantages of UAVs application in engineering management become more and more obvious.Particularly as a non-contact monitoring tool,research on major construction project management based on UAVs monitoring is of significant importance but also faces challenges.On the one hand,decisions about the appropriate scale of charging stations need to be made long before the project starts,affecting the deployment capacity of UAVs.The key lies in how to deploy different quantities and types of UAVs for the charging stations.A reasonable UAVs layout and deployment solution can ensure the practical monitoring requirements at construction sites,maximize the shared use of UAVs,and expand resource utilization.On the other hand,major construction projects have long construction cycles,dynamic changes in construction environments,and uncertainties in tasks at monitored sites at different stages.It is crucial for managers to consider uncertainty factors in scientifically planning the UAVs monitoring routes.Timely adjustments to UAVs deployment under uncertain monitoring conditions can ensure that major construction project management responds to various random situations,ensuring the relative stability of decision solutions.To address the characteristics of the above challenges in major construction management,this study introduces an integrated problem of UAVs deployment and scheduling,considering the uncertainties of monitored tasks at various stages of major construction projects.A two-stage stochastic programming model is proposed to deploy multiple types of UAVs in response to monitoring sites in the context of uncertainty.The objective is to minimize the total cost,including the initial construction cost of charging stations,UAVs startup costs,and flight costs.The first stage of the proposed model concerns decisions on the scale of charging stations and the deployment of multiple types of UAVs.In the second stage,the decision-making involves the UAVs routes for monitoring tasks in different scenarios.By analyzing the interrelationships between decision variables in the two-stage stochastic programming model and combining model features,a tailored two-level heuristic algorithm based on variable neighborhood search and column generation is designed.The outer-level addresses the UAVs-charging station deployment problem in the first stage,applying a variable neighborhood search algorithm for iterative optimization.The inner-level addresses the scheduling problem of UAVs executing monitoring tasks in different scenarios,applying a column generation-based heuristic algorithm for optimization.We conduct some numerical experiments to evaluate the effectiveness of the proposed model and the efficiency of the proposed algorithm.The proposed model and algorithm can be integrated into decision support systems for engineering project management,assisting in optimizing resource allocation at low cost and efficiently completing UAVs monitoring scheduling.This study contributes to the literature in several respects.From a modeling perspective,the proposed two-stage stochastic programming model is comprehensive and combines UAVs deployment for long-term decision-making with UAVs monitoring for short-term decision-making.From an algorithmic perspective,we design a two-level heuristic algorithm based on variable neighborhood search and column generation.Our numerical experiments show that the proposed two-level heuristic algorithm runs efficiently and stably.From a practical perspective,we conduct a series of sensitivity analyses to reveal the influences of detail factors such as the number and the speed of UAVs.The results of our sensitivity analysis reveal some practical managerial insights for engineering project managers.In particular,maintaining a larger number of UAVs can ensure meeting the monitoring requirements at construction sites but may increase unnecessary fixed costs.For project managers,it is necessary to seek a balance between reducing costs and meeting monitoring requirements.Project managers do not need to demand overly fast UAVs speeds,as this does not always lead to cost reduction and may instead increase the cost of purchasing high-speed UAVs.Project managers should determine an appropriate UAVs speed.
major projectunmanned aerial vehicletwo-stage stochastic programmingcolumn generationvariable neighborhood search
万方数据
维普
