Weather-wise: working with the weather to improve construction productivity

The influence of unforeseen, and not just necessarily extreme, weather conditions have adversely impacted Costain's construction productivity causing them significant project delays and economic losses. Costain's smart motorway projects, like most infrastructure projects, are predominantly carried out outdoors and involve multiple weather-sensitive activities. However, different activities are often susceptible to different combinations of weather variables and/or intensities. This has made the analysis of interaction between weather and construction productivity quite challenging so far.

Nowadays, the UK government is committed to investing over £300 billion on an upgraded infrastructure system by 2021. Late projects delivery may have a significant economic impact at the country level. This, as missing the timely exploitation of this new infrastructure during a period of heavy investments can harm the country's economic growth and the financial stability of multiple stakeholders, including the government.

The aim of this project is to tackle the challenge of harnessing the weather seasonal average variation from a construction-relevant perspective. The objective is to develop a holistic and quantitative Operational Research scheduling tool that anticipates the likely occurrence of specific combinations of weather events that prevent the satisfactory execution of frequent construction operations. Not extreme weather events, just sequential phenomena and/or combined non-extreme weather events, which condition over 95% of most contractors' daily operations.

Particularly, the tool will implement a recent Stochastic Operational Research (SOR) model proposed by the investigators, which will allow Costain to optimise the project and activities start dates, their order of execution, even alternate locations across the UK when possible. All these with the aim of shortening the ongoing and future project durations and/or reduce their costs.

Additionally, this project will benefit other construction stakeholders such as public and private project owners. These will benefit from infrastructure projects put in service on time, with more efficient and weather-aware maintenance approaches, while significantly reducing the amount of the frequent weather-related claims.

Regarding deliverables and outputs, the first stage of the project will consist of extending the investigators' SOR model to include smart motorway frequent construction activities. The second stage will involve developing a computer application that can create project schedules linking the former activities to specific (predefined and customisable) combinations of weather variables they are sensitive to. This computer application will implement a set of recent investigators' sine wave regression expressions linking historical weather events probability with the geographical coordinates and day of the year when construction activities can be performed. The third stage will involve the application submitting the initial project schedule to multiple artificially-generated years whose weather will collectively resemble the climatology of the region. In each of those years, the software will calculate the delays each project activity will suffer because of adverse weather. Finally, the computer application will estimate the overall shortest project total duration and the optimum activities order and start dates.

Previous experiments with the current SOR model in building construction have shown that implementing this model at an early stage of a construction project can shorten the project duration by at least 10% (on average), reduce the indirect and overhead costs proportionally and ultimately improve productivity rates. Similar or higher figures are expected for future contractors' infrastructure projects.

Expected dur.: 12 months. Total cost (at 80%FEC): £131,430.90

Keywords: Weather-wise, construction, scheduling, optimisation, Costain, smart motorways

Under a small research grant from the Chartered Institute of Building (CIOB), the investigators recently developed a Stochastic Operational Research (SOR) model that captures and anticipates the manifold interactions between weather and the most frequent construction operations in building construction. The current project, however, will allow Costain (and other stakeholders in due time) to be able to analyse and quantitatively measure the ex-ante and ex-post impacts of the weather in their future and ongoing infrastructure projects.

The model implementation will be achieved by developing a new computer application. This computer application will allow the SOR model to be used by any potential stakeholder who plans, manages, or supervises the execution of any construction work. Specifically, this means that on entering some basic project scheduling information, they will be able to:

1. Calculate how long the same project will take depending on its start date and where their activities are physically located in the UK.

2. Rearrange the order of execution of upcoming activities (for example giving priority to some projects over others) in order to reduce the projects duration and/or costs.

3. Calculate the percentile describing how harsh the weather conditions experienced by a particular project have been when compared to previous years.

4. Modify project start dates and/or work site locations (if not restrained) to find optimum start dates and/or sites distribution that render lower costs and/or project durations.

Outcomes 1 and 2 are of direct interest to construction contractors, and indirectly to project owners. Outcomes 3 and 4 are of primary interest to both of them.

The Investigators' previous small-scale trials in building construction have proved that construction projects may shorten their duration by a minimum average of 10%, with an analogous reduction of indirect and overhead costs.

Infrastructure projects are generally much more exposed to weather than buildings. Hence, on implementing the SOR model, time as well as overheads and indirect cost reductions for infrastructure projects are expected to significantly exceed 10%.

These potential project time and cost reductions have attracted Costain's attention, and they are willing to test the model at a larger scale in their motorway projects. This project constitutes the first step towards the applicants' model scalability.

Whilst there are, of course, some construction works that require forecast information to make appropriate short-term decisions (e.g. pouring concrete today or tomorrow, or anticipate high wind gusts when building an exposed bridge), this proposed project is looking at the applicability of historical climatological and weather information as opposed to pure weather forecasting. Most resource-related operational decisions have to be made with anticipation beyond what current forecasting skill allows. People (workers) have high manoeuvrability, but generally not the equipment, machinery, vehicles and special supplies that they need to work with. As a research translation project the focus is on establishing use of daily historical weather in construction practice whilst recognising potential for future incorporation of sub-seasonal forecasting products.