Management of variations in the public-sector building construction projects in Sri Lanka

Purpose Many public-sector building projects in developing countries are prone to variations owing to the changes made to their original project scopes, and economic depressions and stagnations. Cost and time-overruns are consequences of such variations. Thus, variation management in these projects is important. Hence, this paper aims to analyse the methods of managing variations in public-sector building projects in Sri Lanka. Design/methodology/approach The study adopted a mixed research approach. Seventeen expert interviews and a questionnaire survey were conducted to identify the causes and consequences of the variations, and the strategies that will help manage those variations. Manual content analysis and mean weighted rankings (MWR) were used to analyse the collected quantitative and qualitative data, respectively. Findings The employer, consultant, contractor and several other factors are mostly responsible for the variations in public-sector building projects in Sri Lanka. Cost and time-overruns, rework and demolition and disputes are the major consequences of such variations. The study findings reveal that a detailed employer’s brief, clearly defined project objectives, a comprehensive design and set of specifications, and preliminary investigations are the strategies that will mostly minimise the variations in public-sector building projects in Sri Lanka. Originality/value The study proposes strategies to avoid or minimise the adverse impact of variations, which the industry practitioners can adopt for variation management in public-sector building projects. The study theoretically contributes to knowledge by revealing how variations in public-sector building projects in Sri Lanka can be managed by identifying their causes and consequences.

Squeak and Rattle Prevention by Geometric Variation Management Using a Two-Stage Evolutionary Optimisation Approach

Squeak and rattle are annoying sounds that often are regarded as the indicators for defects and quality issues by the automotive customers. Among the major causes for the generation of squeak and rattle sounds, geometric variation or tolerance stack-up is a key contributor. In the assembly process, the dimensional variation in critical interfaces for generating squeak and rattle events can be magnified due to tolerance stack-up. One provision to manage the tolerance stack-up in these critical interfaces is to optimise the location of connectors between parts in an assembly. Hence, the focus of this work is to prevent squeak and rattle by introducing a geometric variation management approach to be used in the design phase in the automotive industry. The objective is to identify connection configurations that result in minimum variation and deviation in selected measure points from the critical interfaces for squeak and rattle. In this study, a two-stage evolutionary optimisation scheme, based on the genetic algorithm employing the elitism pool, is introduced to fine-tune the connectors’ configuration in an assembly. The objective function was defined as the variation and the deviation in the normal direction and the squeak plane. In the first stage, the location of one-dimensional connectors was found by minimising the objective function in the rattle direction. In the second stage, the best combination of some of the connectors from the first stage was found to define planar fasteners to optimise the objective function both in the rattle direction and the squeak plane. It was shown that by using the proposed two-stage optimisation scheme, the variation and deviation results in critical interfaces for squeak and rattle improved compared to the baseline results.