Most construction projects involve earthmoving operations. Careful management of trucks to perform these operations is considered crucial by earthwork contractors in Taiwan. A mechanism is described that optimizes the management of the truck fleets of earthwork contractors. The proposed model integrates a discrete event simulation technique and genetic algorithms (GAs) to minimize the duration of earthmoving operations. The discrete simulation model is applied to simulate earthmoving operations and generate schedules for dispatching various trucks, and the GAs are used as a filter to screen out schedules that require long trips. Examples prove that the proposed model can determine near-optimal solutions efficiently. In addition, a user-friendly computer program is developed so that earthwork contractors can plan the dispatch schedule of earthmoving trucks easily and effectively.

Most construction projects involve earth removal operations. Careful management of truck fleets to perform these operations is considered crucial by earthwork contractors in Taiwan. This paper presents a hybrid mechanism to optimize the management of the truck fleets of earthwork contractors. We use discrete event simulation models to represent earthmoving operations, and these models are used to simulate schedules for dispatching truck fleets. We then apply genetic algorithms to select the optimal schedules with the least amount of operational time. A theoretical example proves that the proposed mechanism can determine near-optimal solutions efficiently. Finally, we develop a user-friendly interface so that earthwork transportation companies can use the proposed mechanism easily and effectively.

Construction processes are complex operations that include the uses of equipment, materials, crews and geological conditions. Simulation can be used to study these processes in order to make competent decisions that will lower costs and shorten the duration of the project before it has started. Tunneling is a highly repetitive construction operation and simulation can be used to analyze the process. A model used to simulate the NATM tunneling construction operations for Pakuashan highway tunnel project is developed in this paper. This project is unique because of the soil conditions. A discrete event simulation methodology, CYCLONE, is used to build the operation model and simulate the tunneling processes. Statistical analysis indicates that duration of most tunneling operations can be modeled by Beta probability distribution. Results also show the model predicting the advancement rate of such tunneling project in accurate.

Construction accidents are a major contributor to occupational fatalities. To tackle this issue, improved monitoring for hazard elimination is crucial. By introducing a deep-learning image captioning system, we identified hazards via closed-circuit television in construction sites. By leveraging Inception-v3 for feature extraction and the gated recurrent unit for caption generation, real-time hazard monitoring was enabled. Bilingual evaluation understudy scores were determined for continuous and effective hazard detection, helping construction managers enhance safety.

The emerging advances in deep learning and computer vision have enabled traditional cloud-based decision-making through edge computing with artificial intelligent internet of things (AIoT) image sensors (AIoT-IS). As a result, the timeliness and security of image recognition can be obtained. This study aims to develop an AIoT-IS-based smart safety control device for construction sites (“SmartCon” hereafter) for the operations of mobile cranes. The research is carried out to include the definition of hazardous control areas and the identification of unsafe scenarios for crane operations, the development of an intelligent prototype system for the safety control of lifting operations, system application demonstrations, and the evaluation of effectiveness. It is intended to assist labor inspection agencies and industry practitioners with a tool for monitoring lifting operations. The results of two empirical field tests show that the proposed SmartCon improves the safety monitoring of the operation of the machine through the real-time response to the identified potential risks on construction sites.