A Scheduling Model for Linear Projects with Multi-Skilled Resources and Uncertain Activities Time

نوع: Type: thesis

مقطع: Segment: masters

عنوان: Title: A Scheduling Model for Linear Projects with Multi-Skilled Resources and Uncertain Activities Time

ارائه دهنده: Provider: Ali Norouzi

اساتید راهنما: Supervisors: Dr. Mohsen Babaei

اساتید مشاور: Advisory Professors:

اساتید ممتحن یا داور: Examining professors or referees: Dr. Jalal Akbari, Dr. Javad Taheri Nezhad

زمان و تاریخ ارائه: Time and date of presentation: March 6th, 16 untill 18

مکان ارائه: Place of presentation: Amphitheater

چکیده: Abstract: Over the past two decades, effective project planning through optimal utilization of multi-skilled human resources has emerged as a prominent issue. Research findings indicate that increased attention towards the use of multi-skilled models in projects has resulted in significant cost reductions. Being multi-skilled essentially refers to the preparedness of organizational personnel to perform activities beyond their primary responsibilities. In this context, the execution of each activity necessitates a number of human resources possessing diverse skillsets, which must be fulfilled by a subset of project personnel who possess the necessary expertise for the activity. Linear projects, like road construction operations, encounter considerable hurdles with regards to cost and scheduling. The presence of a multitude of uncertainties in the construction process, scheduling, and allocation of human resources has resulted in an escalation of both the duration and cost required to complete such projects. Hence, the development of an efficient scheduling plan and the allocation of skills in the face of uncertainty can have a substantial impact on the progression of project goals. To this end, a mixed-integer linear multi-objective scheduling model was introduced in this research, specifically designed for linear projects that involves human resources with multiple skills. Using the Normalized Normal Constraints (NNC) and TH techniques, the multi-objective model was solved, and preliminary outcomes were presented with GAMS software, using real-world data from a road construction project. Additionally, the study investigated the process of converting the model into a deterministic form and solving it while taking time as a fuzzy parameter. The outcomes of both solution techniques were presented under optimistic, realistic, and pessimistic scenarios, considering non-deterministic activity time. Notably, the proposed model is characterized by its adaptability to various states of fuzzy problem data (i.e., optimistic, realistic, and pessimistic) and the ability to determine the optimal final cost and time of the project based on such states. Ultimately, based the nature of the data and the chosen state, an optimized scheduling plan was presented, comprising of the start and end times of activities within each interval, the quantity of employed human resources, and the competencies utilized within each interval. To comprehend the influence of dividing the project duration on the ultimate cost and duration, the problem model was solved with two segments and single-skilled and multi-skilled human resources. However, no solution was obtained for the problem in both cases. Furthermore, to evaluate the effect of multi-skilled human resources, the problem model was solved by considering four segments and single-skilled human resources. However, no solution was obtained for the problem in this case. Hence, it can be inferred that in large road construction projects, segmenting the project and utilizing multi-skilled human resources are indispensable factors. Moreover, while comparing the TH method to the NNC method for the studied project, the results obtained from the TH method indicate a 63% increase in human resource deployment cost, a 33.4% reduction in project completion time, and a 36.6% decrease in penalty for using human resources with low skill levels, considering the most probable non-deterministic time data compared to deterministic data. Additionally, the final project time, using the multi-objective model, was found to be 4% and 11% less than the results obtained from the LSM method in the deterministic and non-deterministic data scenarios, respectively

فایل: ّFile: Download فایل