The volume of data generated in the world is growing exponentially, with increased usage of the internet and other forms of technologies like mobile phones, the internet of things (IOT), and artificial intelligence (AI). This explosion of data has led to the birth and growth of the buzzing technology referred to as ‘Big data’. Big data is essentially a collection of data that is huge or complex to be handled by traditional tools of data management.
As in other industries, construction has enormous volumes of data. Typically, data management tools have handled structured data like financial documents, project schedules, CAD designs, and employee information. However, these tools fail when it comes to managing the increasing amount of unstructured data in the industry. Collected data can be characterised into:
· Velocity: The speed at which data is generated, captured, and analysed makes a big difference in business results. Big Data tools enable an analysis of data in real-time, therefore, improving the accuracy and timing of business decisions.
· Volume: A construction company may have to handle vast volumes of data from dozens of sources at once, putting pressure on database resources; therefore, calling for Big Data applications that can manage these levels.
· Value: Having lots of data does not necessarily impact the performance progressively. The most important thing is to derive insights from the data.
· Variety: The data collected in construction is generated in various formats and originates from diverse data sources. The different data types change in importance from one project to another.
· Validity/Veracity: Due to the diverse nature of the data collected, a Big Data analyst has to verify whether the data is accurate or credible enough to be used in generating insights.
When raw data is collected, it is prepared through the elimination of incomplete and incorrect data. High-quality data is then entered into its destination, typically a cloud-based system. The future of big data lies in cloud storage, to store and process large datasets. The data is then processed using machine learning algorithms, which converts data to machine-readable formats. The next step is data interpretation, the data presentation in the form of images, plain text, graphs, etc. Finally, information is stored for future or alternative use. In construction, data is analysed for three main reasons:
· Descriptive analysis: This type of analysis involves looking at data statistically to understand what is currently happening. An example, is the use of geolocation and movement data collected from workers’ wearable devices, alongside worker activity data, to study how time is wasted due to movement of people, tools, and construction materials.
· Predictive analysis: In this case historical data is used to generate a model that forecasts what will happen in the future. For example, data collected from a construction site can be used to spot trends and identify factors that lead to accidents, thus improving the safety of workers and property.
· Prescriptive analysis: Once you have conducted descriptive and predictive analysis, the next step is providing recommendations. Using the example from the descriptive analysis, the data collected was used to make adjustments that improved time management and enhanced productivity by an hour per worker a day.
Production of data is continuously expanding, and we even run the risk of too much data. The data we collect in any industry will inevitably be useful in the built environment. Why? People are continually interacting with construction elements: a building, a bridge, a railway, or an airport. Big data is playing a significant role in planning and decision making in Construction; understanding this will go a long way in development.
