Presented By O’Reilly and Intel AI
Put AI to work
Sep 4-5, 2018: Training
Sep 5-7, 2018: Tutorials & Conference
San Francisco, CA

Machine learning for optimizing construction

Ramzi Roy Labban (Consolidated Contractors Company (CCC))
4:50pm-5:30pm Friday, September 7, 2018
AI in the Enterprise
Location: Continental 1-3
Secondary topics:  Temporal data and time-series, Transportation and Logistics
Average rating: *....
(1.00, 1 rating)

Who is this presentation for?

  • C-suite executives, decision makers, managers, and technologists

Prerequisite knowledge

  • A basic understanding of AI and machine learning

What you'll learn

  • Explore the value of applying machine learning for business process optimization


The substantial magnitude of earthworks operations on large construction projects, coupled with uncertainties during estimation and execution, may lead to significant potential delays and losses. Advanced earthworks simulation can help mitigate the associated risks during project estimation and offer project teams better controls during execution.

CCC’s Earthworks Simulator allows engineers to simulate earthworks operations during the various phases of a project, including estimation and execution, as well as retrospectively for lessons learned or claim quantification and justification. The simulator aids estimators and site engineers in forecasting and optimizing equipment crew mixes during the estimation and execution phases of a project. It takes into consideration site technical factors and conditions that affect productivity on earthworks operations, such as equipment condition, operator efficiency, truck cycle times, and material factors.

One of the main parameters that affect truck cycle times is the assumed truck travel speed while hauling material and returning. In planning an earthworks operation requiring material hauling, truck speed is often assumed a certain value based on know-how and previous experience, then fed as a parameter into the earthworks simulator. Any error or inaccuracy in the truck speed assumption will lead to errors in the simulated total cycle time for the trucks. Higher speeds will lead to shorter truck cycle times, in turn underestimating total truck requirements, while lower speeds will lead to longer truck cycle times, in turn assuming the need for more trucks than are actually required.

Ramzi Roy Labban details how CCC uses machine learning—leveraging existing historical truck performance data, including important factors such as construction project size and type, geographical location (and jurisdiction), size of truck, type of truck, time of day, and distance traveled—to predict truck speeds on future operations.

Photo of Ramzi Roy Labban

Ramzi Roy Labban

Consolidated Contractors Company (CCC)

Roy Labban is the director of computer modeling and simulation in the Information Systems Department at Consolidated Contractors Company (CCC), which is ranked among the top 20 international contractors in 2017 by ENR. Roy has 20+ years of experience in software engineering and database application development, business intelligence and analytics, and computer modeling and simulation. Roy is the cofounder and managing partner of a boutique consulting firm focused on delivering business intelligence and analytics for higher education enrollment management. Roy is also the founder and director of a postgraduate coding bootcamp diploma program focusing on new technologies such as the blockchain, artificial intelligence, machine learning, and mobile apps. Roy serves as a member of the Industry Advisory Board of the Computer Science Program (ABET Accredited) at the American University of Science and Technology. He is also a part-time university instructor teaching graduate level courses on computer simulation and machine learning. Roy holds a PhD in construction engineering and management from the University of Alberta and a BE in computer and communications engineering from the American University of Beirut.