Login | Register

Models for Efficient Automated Site Data Acquisition

Title:

Models for Efficient Automated Site Data Acquisition

Ibrahim, Magdy (2015) Models for Efficient Automated Site Data Acquisition. PhD thesis, Concordia University.

[thumbnail of PDF/A version]
Preview
Text (PDF/A version) (application/pdf)
Magdy_Thesis_final_July14.pdf - Accepted Version
Available under License Spectrum Terms of Access.
9MB
[thumbnail of Ibrahim_PhD_S2015.pdf]
Preview
Text (application/pdf)
Ibrahim_PhD_S2015.pdf - Accepted Version
13MB

Abstract

Accurate and timely data acquisition for tracking and progress reporting is essential for efficient management and successful project delivery. Considerable research work has been conducted to develop methods utilizing automated site data acquisition for tracking and progress reporting. However, these developments are challenged by: the dynamic and noisy nature of construction jobsites; the indoor localization accuracy; and the data processing and extraction of actionable information. Limited research work attempted to study and develop customized design of wireless sensor networks to meet the above challenges and overcome limitations of utilizing off-the-shelf technologies.
The objective of this research is to study, design, configure and develop fully customized automated site data acquisition models, with a special focus on near real-time automated tracking and control of construction operations embracing cutting edge innovations in wireless and remote sensing technologies. In this context, wireless and remote sensing technologies are integrated in two customized prototypes to monitor and collect data from construction jobsites. This data is then processed and mined to generate meaningful and actionable information. The developed prototypes are expected to have wider scope of applications in construction management, such as improving construction safety, monitoring the condition of civil infrastructure and reducing energy consumption in buildings.
Two families of prototypes were developed in this research; Sensor Aided GPS (SA-GPS) prototype, which is designed and developed for tracking outdoor construction operations such as earthmoving; and Self-Calibrated Wireless Sensor Network (SC-WSN), which is designed for indoor localization and tracking of construction resources (labor, materials and equipment). These prototypes along with their hardware and software are encapsulated in a computational framework. The framework houses a set of algorithms coded in C# to enable efficient data processing and fusion that support tracking and progress reporting. Both the hardware prototypes and software algorithms were progressively tested, evaluated and re-designed using Rapid Prototyping approach. The validation process of the developed prototypes encompasses three steps; (1) simulation to validate the prototypes’ design virtually using MATLAB, (2) laboratory experiments to evaluate prototypes’ functionality in real time, and (3) testing on scaled case studies after fine-tuning the prototype design based on the results obtained from the first two steps.
The SA-GPS prototype consists of a microcontroller equipped with GPS module as well as a number of sensors such as accelerometer, barometric pressure sensor, Bluetooth proximity and strain gauges. The results of testing the developed SA-GPS prototype on scaled construction jobsite indicated that it was capable of estimating project progress within 3% mean absolute percentage error and 1% standard deviation on 16 trials, in comparison to the standalone GPS which had approximately 12% mean absolute percentage error and 2% standard deviation. The SC-WSN prototype incorporates two main features. The first is the use of the Kalman filtering and smoothing for the RSSI signal to provide more stable and predictable signal for estimating the distance between a reader and a tag. The second is the use of a developed dynamic path-loss model which continually optimizes its parameters to cope with the dynamically changing construction environment using Particle Swarm Optimization (PSO) algorithm. The laboratory testing indicated the improvement in location estimation, where the produced location estimates using SC_WSN had an average error of 0.66m in comparison to 1.67m using the raw RSSI signal. Also the results indicated 60% accuracy improvement in estimating locations using the developed dynamic model. The developed prototypes are not only expected to reduce the risk of project cost and duration overruns by timely and early detection of deviations from project plan, but also enables project managers to observe and oversee their project’s status in near real-time. It is expected that the accuracy of the developed hardware, can be achieved on large-scale real construction projects. This is attributed to the fact that the developed prototype does not require any scalable improvements on its hardware technology, nor does it require any additional computational changes to its developed algorithms and software.

Divisions:Concordia University > Gina Cody School of Engineering and Computer Science > Building, Civil and Environmental Engineering
Item Type:Thesis (PhD)
Authors:Ibrahim, Magdy
Institution:Concordia University
Degree Name:Ph. D.
Program:Civil Engineering
Date:13 July 2015
ID Code:980192
Deposited By: MAGDY OMAR IBRAHIM
Deposited On:27 Oct 2015 19:36
Last Modified:18 Jan 2018 17:50
All items in Spectrum are protected by copyright, with all rights reserved. The use of items is governed by Spectrum's terms of access.

Repository Staff Only: item control page

Downloads per month over past year

Research related to the current document (at the CORE website)
- Research related to the current document (at the CORE website)
Back to top Back to top