Instance-Based Model for Predicting Total Fabrication Duration of Industrial Pipe Spools

• Author / Creator

  Petre, Cristian

• Industrial fabrication for modular installations has its own set of challenges that combine the environments of industrial manufacturing and off-site construction. This hybrid execution strategy means that fabricators need to look at both fields and adopt the best tools and techniques. This thesis presents an investigation into the improvement of delivery time estimates of industrial fabrication of a pipe spool fabrication shop in Alberta. The main contribution of the work is in the area of predicting the total fabrication duration to be expected in order to assist fabrication shop management in planning for appropriate workforce availability and material delivery date requirements. In order to address the objective of improving the prediction of fabrication durations, the spool manufacturing process has been modelled for simulation. However, the data required to validate this model was found to be time-consuming and cumbersome to capture at the required level of details. Alternatively, the development of a data-driven knowledge discovery experiment was pursued. The approach employed was to utilize the fabrication information that was already being captured by the manufacturing facility and evaluate it using instance-based classification. In addition, an effort towards the integration of manufacturing tracking and scheduling estimating is presented. This part of the work will ensure that the schedule is not consulted only at the beginning of a project, but throughout its completion. Updating a schedule with live fabrication progress data will allow production managers to update their completion date estimates and adjust the manufacturing plans to reflect existing issues such as material delivery and labor shortages or performance.

• Subjects / Keywords

  • Pipe Spool Fabrication
  • Instance-Based Classification
  • Construction Simulation
  • Manufacturing
  • Data Classification
  • Process Modelling
  • Nearest Neighbour
  • Industrial Fabrication
  • Data Mining

• Graduation date

  Spring 2016

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/R39Z90P2N

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.