Pandey, P (2019) Lean scheduling in construction projects: realistic milestone planning. Unpublished PhD thesis, Indian Institute of Technology Delhi, India.
Abstract
The efficient construction industry is essential for socio-economic development. The delay in the desired infrastructure not only leads to cost escalation but severely affect the outcome of the dependent socio-economic ventures. To avoid the delay, construction projects are conventionally modelled as a network of conversion processes or work packages (WPs) and scheduling techniques are used for their coordinated execution. These techniques are top-down passive approaches because what happens inside a WP and their linkages are abstracted. The resources and directions are ‘pushed’ downwards without considering execution capability. Due to these unrealistic assumptions, the initial schedule becomes obsolete soon after the commencement of work. The reactive system of project control creates further confusion, leading to time cost overrun. Lean Construction is a bottom-up approach based on ‘TFV Theory’ of production. It enhances value for the customer through optimization and/or elimination of flow activities in addition to optimization of conversion activities. Last Planner is an established ‘pull’ based ‘lean production system and control’ which attempts to reduce inflow variations through commitment-based execution plan and enhances efficiency through continuous improvement. The Last Planner primarily uses two key inputs; 1) Milestones from the master schedule to prepare the phase schedule by planning backwards and; 2) Feedback from the workers for the weekly plan. Since the master schedule is generally based on unrealistic assumptions and there exist several uncertainties associated with workers’ (in terms of availability and capability), the implementation of lean in construction becomes questionable. Experts opined that true benefits can accrue only if the master schedule is realistic and adequate number of workers with requisite skills are available. Thus, main objective of the present study is to develop a robust proactive ‘Lean Scheduling Framework’ which reduces project duration, ensures optimum workers utilization and provides realistic master schedule to enhance the performance of ‘Last Planner’. The proposed solution involves realistic modelling of the construction project using combination of ‘network of main conversion activities’ and ‘feeding processes’ followed by realistic scheduling. The scheduling process involves series of schedule iterations using project specific data to progressively enhance schedule reliability, synchronization of the activity execution rate by modifying workers employment (i.e. using multiple crews, crew multitasking, crew resizing, etc.) for continuous flow, etc. The final schedule obtained after iterations has been termed as ‘Lean Schedule’ because scheduling framework is based on ‘Lean’ concept and provides ‘reliable schedule’ to augment the LPS (Last Planner System). Initially, the components of proposed lean scheduling framework were derived from the literature and integrated using logical reasoning followed by testing of these components were tested using pilot case studies. The findings from the case studies were utilized to progressively improve the framework. Finally, entire framework is tested using data from an ongoing mass housing project. The pilot studies and verification cases revealed that the framework is too generic and can be applied to most of the construction project. It was observed that the accuracy of the lean schedule largely depends on the quality of data utilized, and capability of expert involved in modelling and imitation of project conditions.
Item Type: | Thesis (Doctoral) |
---|---|
Uncontrolled Keywords: | accuracy; continuous improvement; cost overrun; duration; optimization; reasoning; reliability; skills; construction project; employment; feedback; lean construction; project control; scheduling; variations; case study |
Date Deposited: | 16 Apr 2025 19:35 |
Last Modified: | 16 Apr 2025 19:35 |