Keywords

5.1 Introduction

As in many industries, the construction sector has been impacted and somewhat changed by the growing volume of heterogeneous data available at different stages of the construction process. This trend is expected to continue as technologies such as sensors and the Internet of Things (IoT) become more and more accessible and commoditised. The availability of data is particularly useful in the context of Deep Renovation (DR)Footnote 1 where it can dramatically accelerate the decision-making for building stock retrofit. This chapter defines Big Data (BD) and analytics in the context of DR and describes how the use of BD and advanced analytics such as Machine Learning (ML) and Artificial Intelligence (AI) may impact different stages of the DR life cycle.

The remainder of this chapter is organised as follows. Section 5.2 provides an overview of BD and different types of analytics. Section 5.3 presents a series of use cases and applications of BD in construction. Section 5.4 describes how BD can be used in the DR space. Section 5.5 discusses the advantages and benefits of BD in the context of DR. Section 5.6 outlines challenges and barriers related to the adoption and use of BD in construction generally and in DR more specifically. Section 5.7 presents potential future developments and, finally, Sect. 5.8 contains some concluding remarks.

5.2 Big Data Analytics

BD analytics deals with large, heterogeneous data sets from various sources. Data-driven decision-making entails finding trends, patterns, and correlations in data. In order to do that, different types of BD analytics can be implemented which can be classified under four main categories—that is, descriptive, diagnostic, predictive, and prescriptive analytics. Data mining, cleansing, integration, and visualisation enable data analytics in various domains and change and/or improve DR processes, delivering commercial and societal benefits (Rajaraman, 2016; Koukaras & Tjortjis, 2019; Kousis & Tjortjis, 2021).

Descriptive analytics is a popular way for organisations to analyse current and past trends and operational performance. It is the initial stage in interpreting raw data by applying relatively basic statistics and creating sample and measurement statements.

Diagnostic analytics is a type of BD analytics used to evaluate data and content. This form of analytics typically answers questions like ‘why did something happen?’ and therefore aims to explain the causes behind particular results.

Predictive analytics involves estimating outcomes using data insights. It typically employs ML and statistical modelling to predict the most likely outcomes.

Prescriptive analytics is built on the insights gained from descriptive, diagnostic, and predictive analytics to optimise operational processes using simulations and related tools (see Chap. 4 for more details). It uses statistics and data modelling to assist organisations understand and predict the market or environment. It helps individuals define priorities and recognise what might lead to financial or other types of rewards.

5.3 Use Cases and Applications of Big Data in Construction

BD analytics is backed by BD engineering, which has significant construction applications. BD engineering involves Building Information Modelling (BIM)Footnote 2 to enhance project management (Huang, 2021), building design and monitoring performance (Loyola, 2018), safety, energy management, decision-making design frameworks, resource management (Ismail et al., 2018), quality management, waste management (Wang et al., 2018), and more (see Chap. 3 for a more detailed discussion). Moreover, BD platforms that perform BD analytics in construction are essential to BD engineering and may be classified as Horizontally Scaling Platforms (HSPs) and Vertically Scaling Platforms (VSPs). HSPs use several servers by spreading processes and adding additional devices, while VSPs scale by updating the server’s hardware. Waste management (Bilal et al., 2016b), profitability performance measurement (Bilal et al., 2019), smart road construction, and others (Sharif et al., 2017) typically employ HSPs, while VSPs have been mostly used in construction (Curtis, 2020) and transportation (Shtern et al., 2014).

Furthermore, deep learning–based flood detection and damage assessment (Munawar et al., 2021), project delay risk prediction (Gondia et al., 2020), construction site safety (Tixier et al., 2016), construction site monitoring (Rahimian et al., 2020), and neural network models to predict concrete qualities (Maqsoom et al., 2021) are a few instances of AI and ML in construction.

5.4 Big Data and Deep Renovation

The fundamental components of BD engineering include both distributed and parallel processing. BD analytics has been used in the construction industry for a variety of purposes, including waste management (Lu et al., 2016), management of prefabricated building projects (Han & Wang, 2017), profitability studies, and other construction management applications (Bilal et al., 2019).

BD in construction uses AI and ML for revitalising sustainable architecture, energy-efficient building design, and minimising environmental and climatic consequences. Recent advancements in internet speed, accessibility, processing cost, and data storage cost make BD a vital AI supplement (Mehmood et al., 2019).

In recent years, AI has contributed significantly to improving learning-based decision-making. Its use in building design and engineering along with BIM is offering new options for DR utilising BD since very large volumes of construction-related data are available. DR is one of the main drivers for Greenhouse Gas (GHG) emission reduction in cities (Avramidou & Tjortjis, 2021) and along with ML and AI introduces new design potentials, constraints, and solutions. Overall, BIM and Industry Foundation Classes (IFC) improve DR’s decision-making and the energy efficiency of retrofitted buildings (Mulero-Palencia et al., 2021).

Nowadays, DR should aim to harness the maximum economic energy efficiency potential of construction activities at a large scale, thus utilising BD for construction purposes. It should also concentrate on improvements of the building shell of existing structures, leading to extremely high-energy performance. Nonetheless, residential efficiency improvements or criteria (e.g. shell upgrades or Heating, Ventilation and Air Conditioning (HVAC) and hot water system upgrades) vary by climate (Cluett & Amann, 2014).

Despite the European Union’s (EU) energy efficiency targets and renovation actions such as aesthetic improvement of the building outer façade, increased thermal comfort and energy efficiency, and CO2 emission minimisation, the construction industry has not yet adopted large-scale standardised retrofitting techniques that would involve BD analytics in construction (Glumac et al., 2013). Most renovation options include external/internal insulation, air tightening the transparent and opaque building envelope, roof conversion, solar panels, heat recovery, and more efficient HVAC systems. Conventional energy retrofits focus on single system upgrades, such as façade, lighting, and HVAC equipment, without considering integrated renovation options.

5.5 Advantages and Benefits

Literature suggests a number of opportunities for BD adaptation in the DR context (Bilal et al., 2016a):

  1. 1.

    Generative design. The idea is to automate the development of several design models based on specific objectives such as functional requirements, material type, manufacturing process, performance standards, and cost limitation. Such tools use advanced algorithms to develop design solutions that fulfil design criteria. Designers evaluate the designs’ performance and are able to change design objectives and restrictions until they are satisfied.

  2. 2.

    Clash detection and resolution. BIM models should identify design incompatibilities. For effective project management, this step should come before construction. Traditional paper-based procedures, which are less efficient and accurate in identifying design issues, are being replaced by BIM-enabled automated techniques. Design conflict detection involves time-consuming non-trivial design exploration strategies. BD technology may improve knowledge representation and processing via distributed and parallel computing.

  3. 3.

    Performance prediction models. These models employ a vast number of variables and their combinations, which affect each other and overall model performance. They are implemented utilising basic statistical approaches or more complex computational methods such as artificial neural networks. Therefore, these systems involve a large number of variables, something that is computationally intensive, time-consuming, and difficult for existing technologies to perform in real-time. BD technologies may improve real-time processing, model creation, and visualisation.

  4. 4.

    Visual analytics. Analytical issues may be categorised as the ones that require logical solutions and the ones that require heuristic solutions. The first can be automated, whereas the latter are tackled by proper visualisation. Effective visualisation requires human expertise, imagination, and intuition; thus, human knowledge works well with smaller data sets but not with high-dimensional data sets. Visual analytics combines automated reasoning and graphical representations to address complicated analytical issues and require BD to visualise data for enabling personal viewpoints and interactive data exploration.

  5. 5.

    Social networking. The majority of construction sector issues are communication-related. Social media can pose as a fascinating development that might help businesses promote good communication among project teams. Communication via social media is a tendency that has been steadily infiltrating the business sector. The next application areas might include social networking platforms for sharing updated project information, as well as other initiatives for conveying the best sustainability strategies/practices. Yet, strong frameworks need to be conceived to capture all valuable social interactions in BIM forms, from initial design to the final model. Since social media data are likely to be varied, rapid, and massive, BD may be used to construct novel domain applications to promote stakeholder productivity.

  6. 6.

    Personalised services. Such services emphasise on adapting facilities to user preferences. Users control how services are utilised and these systems adapt to user behaviour. They consider both human and automated input. Therefore, personalisation solutions monitor the surroundings for occurrences of interest, creating vast amounts of data. BD technologies can analyse these data streams in real-time to create actionable insights for nearly instant adaption. To do so, contemporary buildings need BD-enabled platforms with a uniform interface to facilitate such personalisation services.

Other advantages/benefits of BD and analytics in construction include:

  1. 1.

    Improvements in construction efficiency. By delivering clear, comprehensible data and detecting possible structural flaws before they occur, data analytics technology aims to cut construction time and material costs. This enables project managers to make faster and more informed choices, thus reducing human errors (Lynn et al., 2021).

  2. 2.

    Environmental impact reduction. Integrating BD that is actually historical project construction data can be blended into BIM technology to precisely estimate the materials and energy required for upcoming projects. This cuts down on unnecessary building waste and enables planners to explore more options for energy-efficient solutions when feasible (Androutsopoulos et al., 2020).

5.6 Challenges and Barriers

In recent years, energy efficiency has become one of the EU’s top priorities (Koukaras et al., 2021a). Some generic barriers of BD in DR are (Lynn et al., 2021) as follows:

  1. 1.

    Human. Several variables may impede the approval, support, and adoption of energy-efficient behaviours, technologies, and initiatives. Social norms, behavioural patterns, inability to use new technologies, lack of information on energy consumption and energy-saving opportunities, and more are all barriers. Moreover, education, age, and family composition affect the adoption of energy-efficient equipment. All these underscore the necessity of adjusting communication to various groups and educating construction experts for adopting and using BD analytics for analysing these data to elevate DR.

  2. 2.

    Technological integration. DR when supported by BD involves multiple domains, stakeholders, and technologies. Interoperability improves communication, coordination, cooperation, collaboration, and distribution in DR projects. This causes interoperability concerns that hinder data flows and value creation when BD is involved. Linking data throughout a restoration project’s lifespan offers a variety of obstacles such as detecting and reconciling disparate schemas and object representations, incompatibilities across data sources, incompatible levels of abstraction, and data quality concerns.

  3. 3.

    Organisational. DR demands senior management commitment with interdisciplinary skills, time, expenditure, skilled personnel, and appropriate technical infrastructure. The absence of appropriately qualified energy efficiency specialists, data scientists, and construction workers in the right selection and installation for integrating new constructional technologies is a major obstacle for DR especially when supported by BD analytics. Project delays and interruptions, sub-optimal energy efficiency, and failure to achieve expected cost reductions as well as high initial investment costs, finance availability, and payback time are also obstacles.

More barriers related specifically to data, that is, BD, applicable in the context of DR are (Bilal et al., 2016a): (a) data security, privacy, and protection, (b) data quality of construction industry data sets, and (c) fast and reliable internet connectivity for BD applications.

In addition, other challenges related to BD handling in construction are (Yousif et al., 2021): (a) inefficient BD experts/data collectors, analysts, and presenters along with the dynamic nature (e.g. online data streams) of BD databases; (b) high expenditure in BD infrastructure/experts, which will prevent enterprises from adopting BD technologies; (c) governments and corporations, which avoid sharing important data with the world, thus forcing data protection policies to be established.

Furthermore, another study specifically looks at BD for energy efficiency in building and notes data access challenges (Marinakis, 2020).

Finally, possible barriers are also related to social and environmental aspects in the context of BD. Aside from reducing energy consumption (Koukaras et al., 2021b), building renovations are typically motivated by issues such as structural repairs (D’Agostino et al., 2017). Buildings utilise 38% of EU energy and produce 36% of CO2. For example, the Dutch non-profit building stock’s DR ratings for 2010–2014 were based on the energy performance of 850,000 homes. The data were obtained from a system that monitored 60% of the sector’s buildings. Despite renovations, the dwellings’ energy efficiency did not alter much (Filippidou et al., 2017).

5.7 Future Developments

BD integration potentially benefits construction companies and all the other stakeholders involved at different stages of the DR life cycle. Using BD for business and environmental sustainability offers construction companies major prospects. BD may help the building sector overcome present hurdles. Using historical and current project data may assist in fostering long-term infrastructure. BD in construction helps prevent errors and yield better construction outcomes.

Future studies could investigate the integrated data that will be utilised for worldwide commercialisation of BD analytics for DR. This involves developing web/mobile applications that can be linked to BD integration systems to show real-time data analytics at a low cost, as well as work on the data-gathering process in the construction fields (Yousif et al., 2021).

Furthermore, future work foresees aspects related to (a) construction waste simulation tools, (b) BD analytics that enables linked building data platforms, (c) BD-driven BIM systems for construction progress monitoring, and (d) BD for design with data (Bilal et al., 2016a).

Future construction research will depend largely on BD since it can help develop better infrastructure and building designs. Construction must automate and integrate technologies to make BD utilisation simple and easy. BD technologies, BIM, and Computer-Aided Design (CAD) cannot be used without proper support and integration. The building industry’s future rests on steadily improving the current conditions (Gbadamosi et al., 2020).

Finally, BD is essential to future building DR projects and data are essential for establishing training models and facilitating construction in general. Future improvements in this area will involve additional algorithms and models that depend on BD for reliable training.

5.8 Conclusion

The objective of this chapter was to expose the reader to the concepts of BD and DR. Technologies, such as prefabricated exteriors, ICT-support for Building Management Systems (BMS), incorporation of Renewable Energy Systems (RES), BIM and building performance simulation models, and high-tech HVAC systems, are just not enough for reaching EU climate change policy goals by 2050. There are still open issues for future innovation in order to conceive effective policies and suggestions for DR implementations (D’Oca et al., 2018). Thus, BD analytics can be employed for the construction sector and more specifically in the context of DR passing on the discussed advantages and benefits.

Nonetheless, the building sector has not yet fully embraced BD. The fast rise of this technology over the past two decades has increased the number of models and platforms for digitising diverse areas. The literature reveals several resources and platforms that may be used for construction management. Yet, currently there is poor adoption in DR and the building business must use and commercialise BD.

Future developments will benefit from internet tools and technologies that allow infrastructure modelling and CAD. These relate to the implementation of efficient energy measures, any prospects for climate change mitigation, and better management for thermal comfort in the context of BD and DR. Simple and inexpensive renovations frequently miss the opportunity to save more energy at a reduced cost. Any DR initiative should include many locations with different building, regulatory, market, and climatic conditions, thus involving BD.

The importance of BD and analytics for enhancing DR was highlighted using representative paradigms from the recent state-of-the-art. Social, economic, and environmental perspectives were also taken into account. In order to make the most out of the large amount of information accessible in the current BD environment, new analytical skills for DR must be developed.