Thomas K. L. Tong

Non-structural fuzzy decision support system for evaluation of construction safety management system

Abstract Because of the poor safety records, there have been a number of safety improvement measures developed in the construction industry of Hong Kong. However, not all of these programs are cost, time and resources effective. If the degree of effectiveness of these programs can be compared and analyzed, it helps management focus their effort and better deploy resources. This study tries to evaluate the safety management systems and prioritize these measures with the consideration of various decision criteria. The Non-structural fuzzy decision support system (NSFDSS) is applied to facilitate the decision making process for these multi-objective problems. Modified NSFDSS is presented that is suitable for the appraisal of complex construction problems, which allows assessment based on a pair-wise comparison of alternatives using semantic operators, even under the condition that insufficient precise information is available.

Site pre-cast yard layout arrangement through genetic algorithms

The use of modular construction has gained wide acceptance in the housing sector. Standardized modular units are often pre-cast on site. The establishment of site pre-cast yard, in particular arranging the pre-cast facilities within the compound, presents real challenge to site management. This complex task is further aggregated with the involvement of several resources with different transport cost. A GA-model is developed for the search for a near optimal layout solution. The fitness function is to minimize the total transport cost for a pre-determined daily output. The use of the model is illustrated by an example. When compared with the best solution within the initial population, 18.45% reduction in cost for resources flow was achieved by the near optimal layout arrangement arrived at the 673rd trial. It is also suggested that the model can be extended to other layout problems such as warehouse and production line.

GA-ANN model for optimizing the locations of tower crane and supply points for high-rise public housing construction

Site layout planning is a complicated issue due to the existence of a vast number of trades and inter-related planning constraints. In this paper, artificial neural networks are used to model the non-linear operations of a key site facility: a tower crane — for high-rise public housing construction. Then genetic algorithms are used to determine the locations of the tower crane, supply points and demand points by optimizing the transportation time and costs. The scope of this study confines to a defined area of construction: the structural concrete-frame construction stage of public housing projects. The developed genetic algorithm model for site facility layout and the artificial neural network model for predicting tower-crane operations are evaluated using a practical example. The optimization results of the example are very promising and it demonstrates the application value of the models.

Comparing non-structural fuzzy decision support system and analytical hierarchy process in decision-making for construction problems

Analytical Hierarchy Process (AHP) is a commonly used decision-aiding tool for resolving multi-criteria decision problems. However, users sometimes find it difficult to ensure a consistent pairwise comparison between voluminous decisions. The cause of which is that the Consistency Ratio (CR) is produced after the evaluation process and its global acceptance criteria is limited. When the derived ratio reports some inconsistency, it requires a long process to locate and rectify the problem. The major aim of this study is to look for an alternative decision-aiding tool to AHP, helping to avoid the above problem. The alternative approach proposed in this study is the Non-Structural Fuzzy Decision Support System (NSFDSS). The application of the system is illustrated with a worked example. The results generated by NSFDSS are compared against those generated by the conventional AHP that shows the effectiveness and some unique advantages of the proposed tool over AHP.

Selection of vertical formwork system by probabilistic neural networks models

A good formwork system enables speedy completion of the concrete structure, following which other subsequent trades can be started. However, the current intuitive judgment approach in the selection of formwork systems cannot assure an optimal and consistent result. Artificial neural networks may improve the selection process. Formwork represents a significant part of the cost of concrete structure construction. Most subsequent trades including internal finishing and external cladding depend on the completion of the building structure. A suitable formwork system is thus crucial for maintaining the smooth flow of the various trades and a proper working sequence of various work activities. Based on data collected from a previous study, it is clear that the key factors affecting the selection of a relevant formwork system include building height and structural system, concrete finish, site conditions, availability of equipment and building shape. Neural network models are developed for the selection of vertical formwork systems using the architecture of the probabilistic neural network (PNN) model. A case study verifies the validity of this approach.

ELECTRE III in evaluating performance of construction plants: case study on concrete vibrators

In this paper, we discuss how a multi‐criteria decision‐making technique (ELECTRE III) facilitates construction practitioners in decision making for selecting an appropriate construction plant. We first introduce the theoretical background of the technique. Then a case study on how to apply the technique to select concrete vibrators is presented. The findings demonstrate the effectiveness of the model in outranking various options in the process of concrete vibrator selection. The use of ELECTRE III can systematically analyse expert judgments on the decision factors and alternatives that can rationalize the selection process.

Rough set theory for distilling construction safety measures

There are numerous construction safety measures adopted by the local construction industry in Hong Kong. The purpose of this study is to distil the more significant measures from all these practices. To achieve this, the rough set theory, a data mining technique by applying the rule induction method, is proposed to distil the rules that determine the safety performance of construction firms. Rough sets represent a different mathematical approach to vagueness and uncertainty. It is based on the assumption that lowering the degree of precision in the data makes the data pattern more visible. Under such an assumption, the rough set theory can provide the ability of classifying vague and uncertain data. A practical example is used to illustrate its application to distil these safety measures and highlight those which are most effective and important in combating site accidents. There are three decision rules identified; the best one is companies with a comprehensive safety orientation programme and good safety award campaigns for senior management staff which give the lowest accident rate and the best safety performance. Safety management rules can be successfully reduced, facilitating contractors to direct their limited recourses effectively.

Conflict analysis study for public engagement programme in infrastructure planning

Purpose – Stakeholder management in the form of public engagement in urban planning and infrastructure project impact assessments have become increasingly prevailing. However, the exercise may involve some complex socio‐economic considerations and critical conflict analysis/resolution scenarios. Without a scientific approach, it is difficult to distil a resolution from the chaotic public opinions, especially in those highly controversial issues. This paper proposes the use of conflict analysis to screen out the resolution in a scientific manner.Design/methodology/approach – This study uses a case of harbor‐front design in Hong Kong to illustrate how the analysis was carried out to accomplish the objectives.Findings – The example has illustrated that the conflict analysis has helped resolve a set of major issues in the Hong Kong harbor‐front redevelopment.Originality/value – The introduction of a structured conflict analysis framework for the government to systematically deal with the conflicting viewpoint...

Genetic algorithm model in optimizing the use of labour

The construction industry is characterized by the existence of multiple trades and crafts. With the existence of multiple-tiers of labour-only-subcontracting in Hong Kong, tradesmen are normally assigned to tasks of a narrowly defined skill. Lately, there has been a call for the adoption of a directly employed labour scheme by the Hong Kong Housing Authority and the public works departments in Hong Kong in order to improve both safety and quality. However, current industry practice has hindered the adoption of directly employed labour, which requires assigning tradesmen to broadly defined task groupings. In implementing the scheme, the first thing to be resolved is how to maximize the levels of use of workers, because that is the major financial incentive to encourage contractors to adopt the directly employed labour policy. This coupled with the shortage of some skilled craft workers call into question the traditionally accepted ‘single-skilled’ or ‘single-task’ approach in labour deployment. A genetic algorithm model is developed to optimize the labour deployment and practical examples are presented. The optimization results are very promising, confirming the practical application value of the model.

An integrated system for earthmoving planning

In earthmoving operations, there are a number of activities that need to be considered; for example, path identification, plant selection, assessing the compatibility of the paths and the plants so selected, cost and productivity comparison of alternatives, safety and environmental considerations, etc. The traditional approach to this problem by heuristic rules generated from the experience of planners cannot guarantee an optimal solution. Hence, a scientific method is proposed to automate the earthmoving planning by integrating: (1) a path‐finding algorithm; (2) a plant selection system; (3) application of extenics theory to address the compatibility; and (4) the use of genetic algorithms to optimize the alternatives in terms of costs, productivity, safety and environmental considerations. The integrated system is illustrated step by step using a genuine construction project as an example. The result demonstrates the effectiveness of the system in automating the earthmoving planning exercise.