AN INTRODUCTION TO BIG DATA IN CONSTRUCTION – WHAT CAN WE EXPECT?
It’s no secret that data is transforming business. Big data is a term that we hear a lot about, and new and innovative uses of data are cropping up every day in every industry. It reflects the massive growth in the quantity of available and relevant data in society, industry and business, and how data can improve decision making.
This report provides some excellent examples. One recent deployment of big data in construction is Caterpillar. Equipment Telematics involves the recording of millions of pieces of daily data around equipment usage and health. To make sense of it all, Caterpillar has partnered with technology startup Uptake, to co-develop predictive diagnostic tools that turn equipment data into meaningful information, identify maintenance issues, and minimize downtime.
The Caterpillar and Uptake partnership is just one of many examples of organizations looking to harness their data for decision-making.
What is Big Data
While it may seem like a relatively new concept, technologists and researchers have been talking about big data for a long time. The term “big data” was first used in a 1997 NASA article that identified the challenge of working with large data sets that taxed the computing power of systems available at the time. For an interesting chronology on big data, read this Forbes article.
Today, the rate of data creation continues to increase exponentially, driven by advances in data acquisition and storage technology, as well as the ubiquity of devices and platforms that generate data. Big data is defined by Gartner as having three dimensions that distinguish it from traditional data sources:
- Variety – a range of data sources, types, and potential uses
- Velocity – speed with which data is created, assimilated and used in some way
- Volume – amount of data being created and stored
Big data also describes the ability to derive insights from the relationships among large data sets, and the way society, science and industry uses this information. It also refers to impacts on organizations and business processes operating with an abundance of data and how to transfer insights into competitive advantage in what is rapidly becoming a data driven market.
Converting Big Data to Insights
If variety, velocity and volume simplifies the definition of big data, Figure 1 below visualizes the functional characteristics of big data at the congruence of information, technology, organizational impacts and methods. Associated with these are a wide range of factors, themes and constructs that further comprise big data.
Figure 1: Constructs & Themes in Big Data¹
Deriving meaningful insights from such a complicated model and converting that knowledge into action is a challenge for any industry, but especially those of us in construction.
Construction is an industry that has historically been slow to adopt technology driven workflows. In a recent survey of global technology leaders, KPMG found that only just 36 percent of engineering and construction firms and 21 percent of owners say they utilize advanced data approaches. The industry clearly has a long way to go.
Harnessing insights from big data requires more than just technology adoption however; firms must also consider organization, culture, and creating workflows and incentives that optimize the use of big data. It is only through this transition that firms can make the transition towards data-driven Value; a fourth dimension, and V, of big data.
Current construction operations rely on simplifying complicated operating environments to create solutions and in doing so, lose visibility of the factors that drive performance. Big data requires measuring, thinking and acting at lower levels of detail, and acting upon the many small events that impact performance. This was described in a recent McKinsey article as transitioning from a knowing culture to a learning culture, where organizations adopt an objective approach to decision-making that embraces the power of data and technology.
Where to Start: Data Collection
So how can the construction industry pursue data-driven change? Knowledge and awareness is improved by better information. Information comprises collections of data. For practitioners of construction management looking to make better decisions, a big data strategy must start with collecting better data.
What does ‘better data’ mean? Your data must have the right attributes. It must be contextually rich, accurate, timely, linked, and digitized. This is complicated in construction where paper-based, manual processes have long been a characteristic of field data capture. To support a data-driven approach to construction management, big data requires firms move to automated processes for digitally documenting project conditions, represented by stage 1 of the data-driven journey outlined in Figure 2 below:
Figure 2: A Data-Driven Journey for the Construction Industry
This foundational step will improve data accuracy, reduce the administrative burden of construction and free up management time for value-added analysis. Solutions like Rhumbix that are designed around data collection for construction projects is one approach that contractors can leverage to kick-start their data initiatives. Additional details on what a data-driven approach in construction involves is further outlined in our white paper on data-driven transformation for construction.
With improved data collection methods, firms can begin to transform their decision-making process to embrace a learning culture driven by a more adaptive, versatile organizational model, where data- driven insights are used to define decision-making in support of organizational objectives.
Successful constructors of the future will challenge the status quo, ask different questions than they did before, and seek out new insights from data. They will start to look a lot more like data companies than they do traditional construction firms and will in turn transform the entire industry.
¹ Andrea De Mauro Marco Greco Michele Grimaldi, (2016), “A formal definition of Big Data based on its essential features“, Library Review, Vol. 65 Iss 3 pp. 122 – 135