Neurofuzzy-based productivity prediction model for horizontal directional drilling

Tarek Zayed, Muhammad Mahmoud

Research output: Journal article publicationJournal articleAcademic researchpeer-review

8 Citations (Scopus)


Productivity prediction and cost estimation of horizontal directional drilling (HDD) as a trenchless technology technique involves a large number of objective and subjective factors, which should be carefully identified and studied. To consider the effect of these factors on productivity prediction, the research presented in this paper assists in developing a productivity model for HDD operations. Potential factors impacting productivity are identified and studied based upon the literature and HDD experts across North America and abroad. A neurofuzzy (NF) approach is employed to develop the HDD productivity prediction model operating in clay, rock, and sandy soils. The merits of this approach involve decreasing uncertainties in results, addressing nonlinear relationships, and dealing well with imprecise and linguistic data. The NF model is tested using actual project data, which showed robust results with average validity percentages of 94.7, 82.3, and 86.7% for clay, rock, and sandy soils, respectively. The model is also used to produce productivity curves (production rate versus influencing factors) for each soil type. An automated user-friendly productivity prediction tool (HDD-PP) is developed to predict HDD productivity based on the NF model. This analysis has proved helpful for contractors, consultants, and HDD professionals in predicting execution time and estimating cost of HDD projects during the preconstruction phase.
Original languageEnglish
Article number04014004
JournalJournal of Pipeline Systems Engineering and Practice
Issue number3
Publication statusPublished - 1 Jan 2014
Externally publishedYes


  • Horizontal directional drilling (HDD)
  • Model
  • Neurofuzzy
  • Productivity

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Mechanical Engineering


Dive into the research topics of 'Neurofuzzy-based productivity prediction model for horizontal directional drilling'. Together they form a unique fingerprint.

Cite this