Simon D. Smith

Estimating the productivity of cyclic construction operations using case-based reasoning

Abstract The estimation of productivity rates in cyclic construction processes is a difficult, but essential task in the planning of construction projects. The conventional method—a calculation by intuition—for devising such estimates is often laborious and inaccurate. To overcome these difficulties this paper proposes that Case-Based Reasoning could be employed to provide estimates from historical cases of construction. To this end a model, Case-Based Estimator (CBE) is presented. CBE is a small model, consisting of five input features, one output and a small case base, and it is suggested that the findings within this paper could be applied to any small model of similar characteristics—not only within the construction field. The development and implementation of the model is discussed and an experiment was performed to assess the ability of two retrieval mechanisms (one a simple mathematical formula, the other an adaptation of the ID3 decision tree generating algorithm) to measure similarity. The simple formula was found to be more preferable, both in terms of consistency and development effort. CBE was validated, not only against the performance of past operations (which were not used in the model development), but also against estimates provided by a professional construction planner. The model was found to provide more precise and consistent estimates than the planner, with 90% of the estimates being within a 10% relative error of the observed value. However, CBE did not perform as accurately when estimating operations which were thought to occur only rarely (outliers).

Rethinking construction cost overruns: cognition, learning and estimation

Purpose – Drawing on mainstream arguments in the literature, the paper presents a coherent and holistic view on the causes of cost overruns, and the dynamics between cognitive dispositions, learning and estimation. A cost prediction model has also been developed using data mining for estimating final cost of projects. The paper aims to discuss these issues. Design/methodology/approach – A mixed-method approach was adopted: a qualitative exploration of the causes of cost overrun followed by an empirical development of a final cost model using artificial neural networks. Findings – A conceptual model to distinguish between the often conflated causes of underestimation and cost overruns on large publicly funded projects. The empirical model developed in this paper achieved an average absolute percentage error of 3.67 percent with 87 percent of the model predictions within a range of ±5 percent of the actual final cost. Practical implications – The model developed can be converted to a desktop package for qui...

Planning, estimation and productivity in the lean concrete pour

With an increasingly competitive global market, the UK construction industry finally realised that in order to survive, a marked increase in efficiency and effectiveness have to be achieved in all areas. This paper will describe the UKs approach to planning and designing the concrete operations that form a major part of many civil engineering construction projects. A productivity study has been carried out on three different construction projects and over 200 concrete pours have been observed. The data and knowledge collected on site have been subjected to lean construction philosophies, producing a “lean” measure of productivity, and it has been shown that major productivity increases could be achieved by implementing several relatively simple principles.

Dealing with construction cost overruns using data mining

One of the main aims of any construction client is to procure a project within the limits of a predefined budget. However, most construction projects routinely overrun their cost estimates. Existing theories on construction cost overrun suggest a number of causes ranging from technical difficulties, optimism bias, managerial incompetence and strategic misrepresentation. However, much of the budgetary decision-making process in the early stages of a project is carried out in an environment of high uncertainty with little available information for accurate estimation. Using non-parametric bootstrapping and ensemble modelling in artificial neural networks, final project cost-forecasting models were developed with 1600 completed projects. This helped to extract information embedded in data on completed construction projects, in an attempt to address the problem of the dearth of information in the early stages of a project. It was found that 92% of the 100 validation predictions were within ±10% of the actual final cost of the project while 77% were within ±5% of actual final cost. This indicates the model’s ability to generalize satisfactorily when validated with new data. The models are being deployed within the operations of the industry partner involved in this research to help increase the reliability and accuracy of initial cost estimates.

Modelling and experimentation of the concrete supply and delivery process

Like many construction processes, concrete delivery and supply is stochastic. This system cannot therefore be modelled deterministically, using average data as input, as the resulting estimates of productivity and cost would be not take into account any negative effects of the random nature of events. The many factors that influence the concrete system can be summarised into : truckmixer interarrival time, truckmixer position time, concrete load pump time, truckmixer volume and concrete rejection rate. The work described in this paper starts with a model of the above factors, based on actual observations of real concreting activities. The random variability of the factors has been incorporated into the model by the use of the gamma distribution, which was found to match the raw data better than other commonly used probability distributions. Of the many methods available to analyse such a model, simulation was used because of its ease of application and because it allows the model to be experimented with: if the model is valid, results obtained through the experimentation will have both significance and use in the management of the real system. The main results of the experimental analysis concern the optimisation of the concreting process: the maximisation of productivity and the minimisation of cost. It is the latter which provides the most interesting and useful conclusions, which shows that the cost of a concreting operation to the concrete supplier varies significantly depending on the operating conditions. This variation in cost is much more marked than the variation of contractor cost- an important anomaly as the contractor has a greater influence over the management of concreting operations.

The selection of risk management techniques using case-based reasoning

Risk needs to be managed at every stage of a construction project and by all organisations involved. Although there is a wide range of techniques available, numerous studies have shown that practitioners rely only on a few techniques. The research presented in this paper has developed a framework for breaking down risk management problems. Assessing 179 examples in the literature against this framework, a library has been created that combines techniques with given problem characteristics. Case-based reasoning (CBR) uses the principle of human reasoning as a methodology for solving problems by matching new cases to similar historic cases. This research has used the similarity measuring capability of CBR to propose risk techniques for new situations. It has been shown through CBR that risk management problems can be defined by a combination of political, economic, social, technological, legal and environmental factors with fuzziness, incompleteness and randomness. A tool has been created that can suggest the most appropriate risk technique to be used for a given situation. Validation has shown that the recommended techniques are the most appropriate, on the basis of the case-base, more than 90% of the time. This work shows the applicability of CBR to the risk management problem and, by exploring the heuristic parameters of CBR, adds to the knowledge base in this type of modelling.

Improving concrete placement simulation with a case-based reasoning input

Discrete-event simulation has been used extensively in the past to analyse construction operations and has been shown to be an effective tool for improving construction process planning. Unfortunately, the widespread application of simulation has been prevented, in part, by the requirement for the developer and user to understand the stochastic features of the process. Further, if the stochastic inputs are not representative of the real system, the simulation output will be misleading. This article proposes that case-based reasoning (CBR) could be used to: improve the input and output data of a simulation model and remove the requirement for a user to have specialist statistical knowledge of the process. Case-based reasoning works by solving new cases from knowledge stored in a case base. Two models are presented in this article: a discrete-event simulation model, MatSim, and a hybrid CBR-simulation model, CBRSim. Both models were developed using real construction data. The models were compared to measure any improvements in simulation output, by using a CBR suggested input. Data from an independent construction project were used to test both models, and the results indicate that CBRSim can achieve estimates of observed values that are more accurate and reliable than those from MatSim.

Identifying Hazards in Transportation Construction and Maintenance Tasks: Case-Based Reasoning Approach Using Railroad Data

Construction and maintenance of the transport infrastructure presents many hazards to workers. In the United Kingdom (UK), safety issues are recognized by government agencies (e.g., Health and Safety Commission, Highways Agency), academia, and industry alike. Increasing skill shortages and an aging working population present problems on collating, organizing, and redistributing safety knowledge before existing workers retire or change jobs. Any influx of international workers also brings problems of language, effective communication, and training into the mix. The knowledge and experience of these workers needs to be tapped and used in the measurement of risks and their subsequent management. In response to these issues, the artificial intelligence methodology, case-based reasoning, has been incorporated into an information technology tool to improve hazard identification and management during a workers daily tasks of identifying hazards and determining appropriate mitigations. The tool prompts users to classify a given work task and then, using a stored library of cases, suggests possible mitigation strategies. The user can accept or reject suggested mitigations. Results are fed back into the system, which becomes more refined for the next work task and next user. The tool is being developed in collaboration with Carillion Transport, a large UK infrastructure development and management contractor. It is in the development stages, but the ultimate aim will be deployment of the tool to those working in the field of construction and maintenance in infrastructure. A working example of the tool is given, followed by a presentation of its strengths and opportunity for improvement.

Exploring Safety Management Challenges for Multi-National Construction Workforces: A UK Case Study

Abstract Large construction projects frequently operate with multi-national workforces, utilizing migrant workers to provide both skilled and unskilled labour. Multi-national workforces are also brought together through joint ventures, as companies from different countries collaborate and share their expertise to construct large and complex construction projects. A multi-national joint venture in the UK provides the case study for an examination of the safety management challenges found on such projects. Whilst language and communication issues amongst workers are typically primary concerns, here they have not been prioritized. Instead, findings are presented that illuminate more nuanced and unquantifiable problems that faced the safety management team. An ethnographically informed approach was mobilized, with the lead researcher spending three years on the site with the safety team gathering data. Analysis revealed several challenges: problems with non-UK company compliance with UK legislation and standards; differences in working practices amongst both non-UK workers and their managers; differences associated with national cultures; and problems of poor worker welfare. It is suggested that awareness of these challenges should inform both the way in which such projects are initially contracted, as well as the development of more sophisticated safety management systems that better support multi-national construction projects in practice.

The health and safety implications of socio-cultural context for community construction projects in developing countries

Community participation in construction during rural infrastructure projects in developing countries is encouraged by many non-governmental organizations. The health and safety aspects of this type of development model have not previously been adequately researched, however. The aim is to identify the socio-cultural factors that motivate community members to participate in construction activities which they perceive as hazardous during a case study of a water and sanitation project in rural Ghana. This is a step towards understanding how health and safety can be more effectively managed during community development projects. A qualitative approach has been taken, using interview, observation and reflection. It was found that the communal culture of the local context resulted in community members feeling pressurized to participate in hazardous construction activities. Local customary laws further compelled individuals as they were concerned they could be fined or arrested should they not fulfil their communal obligations. Further work is required to determine the boundaries within which findings apply but it is likely that there are implications for others managing community construction projects both in Ghana and further afield.