Pricing models of smart buildings in smart environments

Smart buildings as interactions

Classic hedonic pricing models in real estate finance rely on physical structure, location and, in some rare cases illustrated in Chapter 1 of this book on lease characteristics to derive prices of commercial real estate assets. In doing so, models focus on two realms of real estate highlighted by James A. Graaskamp (1981): the space-time realm (physical structure and location) and the money-time realm (lease structure).

In smart environments, there is a third realm of real estate that models should take into account: the digital-time realm which emerges for smart buildings. In their usual formulations, hedonic pricing models do not account for smart buildings’ digital dimension. There are two possible ways to incorporate real estate’s digital-time realm in pricing models:

- First, by adopting a technological approach where the objective measurements of technologies embedded in the built environment (e.g. number of cyber-physical devices in a building) are used, as additional layers of a building’s physical structure.

- Second, by considering a functionalist approach where interactions between buildings and their occupants are assumed to be value drivers in the built environment. Technology underpins these interactions but only as a tool towards a means which, in the context of ubicomp, should be transparent to the end-users.

The first approach does not represent a major conceptual challenge, provided that there are objective measurements available to assess the ‘quantity’ of technology embedded into smart buildings. An industry consensus on these measurements would of course be essential. Conversely, the second approach supposes to develop a taxonomy of interactions in smart space applicable to all buildings (Lecomte, 2019a).

Methods presented in this section rely on the second approach which can be characterised as functionalist. The premise of this approach hinges on a new take on what constitutes a building. In addition to bricks and mortars, technology fuelled space enables interactions between a building and its occupants, thus triggering the emergence of intelligent pervasive spaces. Liu and Gulliver (2013) explain:

[The built environment] provides a context within which social spaces can be constructed, allowing the value of the built environment to be quantified through services and interactions that it provides to users. [...] The building can be seen as a set of designed interaction scripts, which the users evoke when they interact with the space.

Hedonic pricing model of smart buildings based on smart space’s layered structure

The first pricing model adopts an holistic approach to smart buildings’ digital dimension by including into the pricing model the smart environment’s overall contribution to a building’s smartness (i.e. its ability to be smart).

Smartness in buildings is the product of an urban system including the smart grid, ICT infrastructure, and the digital skin. Smart buildings come with “new boundaries” that extend beyond their physical structures and follow smart space’s layered structure. In that sense, smart buildings bring back to the fore pioneer real estate academics’ view on land as multidimensional space underpinning the physical foundation of real estate value (e.g. Fisher and Fisher, 1954). Ratcliff (1961) asserts: “The essence of location derives from one of the elemental physical facts of life, the reality of space”. This was true in 1960s America. And it is still true in smart cities except that the reality of space has changed with the emergence of smart space.

The methodology presented here supposes a holistic approach to commercial real estate value. To encompass the reality of space in smart buildings, it implements a micro-scale analysis of each layer of smart space. According to Lecomte (2019b), smart space is made up of four layers constitutive of physical space (smart grid, building’s physical structure) and digital space (embedded ICT in the built environment, cloud/fog). Figure 2.1 represents the four layers of smart space. Each of these four layers contributes to a building’s smartness.

CLOUD/ FOG: Data Warehousing and Analytics DIGITAL SPACE


PHYSICAL SPACE: Building's physical structure

____________________________ • PHYSICAL SPACE

SMART GRID: Smart prosumer building ’

Figure 2.1 The four layers of smart space after Lecomte (2019a).

To capture the complex synergies among the four layers in enabling smartness in the built environment, Lecomte (2019a) designs an indicator of a building’s smartness after smart space’s layered structure. This indicator is called the Smart Index Matrix (SIM). The SIM is used as an engine to a hedonic pricing model of smart real estate.

• The Smart Index Matrix (SIM)

The Smart Index Matrix (SIM) is a scoring methodology for smart real estate which includes (i) a physical score broken down into a smart grid coefficient and a smart building score and (ii) a digital score capturing digital space’s ICT and data analytics capabilities.

Interactions between the physical score and digital score determine a building’s performances in smart space, which are captured in the SIM (Lecomte, 2019a). The mathematical formulation of the SIM follows a simple matrix equation:

SIM = xA(BC)

where xA is the physical score (x is a scalar parameter capturing the smart grid’s performance and A is a NxK matrix with entries for smart-enabling factors at the building level).

And BC is the digital score (B is a KxN matrix capturing smart-enabling factors in the built environment’s ICT infrastructure and C is a NxK matrix capturing the cloud/ fog’s data warehousing and analytics capabilities). SIM is compatible with the two generic models of smart city: ubiquitous city and augmented city (Lecomte, 2019b). In ubiquitous cities (Aurigi, 2009; Anttiroiko, 2013), matrices B and C are constant irrespective of the building’s location whereas in augmented cities where the digital skin is uneven, B and C are location specific and, possibly, property-type specific.

When adding the digital realm’s micro-scales to commercial real estate analysis, new dimensions of granularity emerge in smart urban environments. As technology becomes “the dominating factor of heterogeneity and the main value driver for commercial real estate in smart cities” (Lecomte, 2019b), pricing models of smart buildings have to account for property heterogeneity stemming from the pervasive implementation of smart technologies in commercial real estate. Technological heterogeneity as a source of smart property heterogeneity materialises at two levels in the SIM: first, at the urban infrastructure level where not all MSAs and neighbourhoods within a city might offer the same quality of ICT infrastructure; second at the level of buildings whose smartness within a given property type and/or geographic area might display great diversity.

Therefore, whilst commercial real estate indices are traditionally segmented by property types and geographic locations, indices of smart buildings can be segmented according to smart space’ layers:

  • - The overall smartness of the building’s physical environment (xA),
  • - The building’s intrinsic smartness based on its physical structure (matrix A),
  • - The overall smartness of the building’s digital environment (BC),
  • - The assessed quality of the ICT infrastructure accessible to the building (matrix B),
  • - The cloud/fog’s data warehousing and analytics capabilities (matrix C).
  • Hedonic pricing model of smart buildings based on the Smart Index Matrix

The premise of the model is that interactions between occupants and a smart building define the property’s value. In smart environments, these interactions are underpinned by smart space’s four layers captured in the SIM. Combinations of constituent characteristics linked to these four layers contribute to the property’s value depending on estimated coefficients indicative of the weights that occupants place on the various dimensions of a building’s smartness. Noticeably, environmental attributes which originate in technology have to be identified and standardised.

This methodology redefines the boundaries of value in commercial real estate by providing a broad vision of property pricing, which is anchored in the complex nature of smart environments’ spatial components. Figure 2.2 illustrates this vision of a smart building in its smart urban environment.

In mathematical form, let Yt be the building’s (In) price at time t such that:

Yt =SIMtyt +et

where yt denotes a Kxl estimated coefficient vector and st is the regression error term.

Hence, property price is directly linked to the four strata of smart space in a model that fully acknowledges smart real estate’s positioning amid an urban ecosystem fostering smartness through synergetic interactions at the property level. The micro-scales involved here are not user-centric, nor experiential but structural in both physical and digital spaces. These micro-scales’ scopes depend on the selection of quality variables in matrices A, B and C reported in the SIM.




Figure 2.2 Smart building in smart urban environment (cloud computing paradigm).

It is assumed that space users have preferences with respect to these smartenabling technological characteristics in a building and its broader environment. It is also assumed that all components as well as their complex interactions can be objectively and reliably measured. A lot of technical expertise will obviously be required in order to set up and effectively run such model.

As smart technologies become more pervasive and keep evolving towards ever more widespread interactions in cities and buildings, quality variables along with their coefficients in the pricing model will have to be regularly re-assessed and calibrated.

< Prev   CONTENTS   Source   Next >