Sallan Snapshot Photo ©amazon pixels


Snapshot Column AuthorSnapshot

Smart Building Technolgy: Not Smart Enough

By: Stephen Samouhos

January 01, 2010

Our carbon-constrained future requires that we improve the energy efficiency of our building stock in order to achieve comfortable spaces at a minimum environmental and financial burden. Smart building technology has been developed and popularized as the key ingredient to achieving this scalable building energy-efficiency, but there are several key challenges to building energy efficiency that remain unresolved by smart building technology. By recognizing and embracing these challenges, we may avoid the pitfalls of a misplaced faith in technology, and generate solutions to the true underlying causes of buildings’ energy inefficiency.

Smart Building Technology
Smart building technology is a suite of software and hardware products and services that use data and information about a building and its environment to automatically optimize the building's operation and maintenance according to user requirements. The technology takes many different forms, but is ultimately enabled by the availability of low-cost sensors and actuators and proliferation of data networking across the urban environment. A smart building has the infrastructure necessary to identify and correct its own sub-optimal operation, adapt to changes in user requirements, and interact with its occupants.

Despite these ostensible innovations that we call smart buildings, much of that same technology has existed and been used in the same way for many years in the manufacturing industry. Our current efforts for wide-scale deployment of information dashboards, automation and data acquisition within buildings follows the same evolution of quality control and automation that happened in most manufacturing industries starting 20 years ago. This time-lag between smart buildings and industrial automation is an artifact of the core difference between the businesses of owning and making buildings versus owning a manufacturing plant; the latter generates profits by virtue of its operations, whereas the former generates profits by the sale and rental of space. Until recently, most building owners did not need to optimize building operations in order to secure a profit, whereas maximizing operational efficiency - thus aggregating reduced costs from all sources - has always been paramount for a manufacturer. Only now is the recent economic downturn in the real-estate market coupled with the rising price of energy forcing real estate owners and managers to change their perspectives and streamline operating costs in order to remain whole. Unfortunately, this change in perspective has come after many generations of building development and the deep-seated past practices of making and owning buildings have resulted in practical factors that drag on the scalable success of smart buildings.

Practical Factors
As the operations of a building become more important to its stakeholders, several critical factors are revealed that are unique to the building industry. Together they present formidable barriers to the proliferation of smart buildings. These include

  1. Lack of as-built drawings
  2. Complex efficiency incentives
  3. Prevalence of custom rather than standardized design/build construction
  4. Limited technical training of service agents
  5. Ubiquitous risk aversion in the industry

Lack of As-Built Drawings
As-built drawings are construction documents that are created after a building is erected; they detail the physical construction of the entire building. The as-built drawing set is especially pertinent to the retro-fit of smart building technology in the existing building stock. A convenient, low-cost retro-fit of smart building technology would utilize as much of the existing building infrastructure as possible, but the available infra-structure is only documented on the as-built drawings. More often than not, and especially for older buildings that constitute the majority of our building stock, there are no as-built drawings either because they were lost, are outdated, or were never created to begin with.

Consequently, smart building technology renovations are almost always plagued by mysteries that lie behind the gypsum board wall; hidden junction boxes with surprising wire splices, pipes that unexpectedly tee off and go to another room, duct work that just ends, and countless other examples. A major obstacle to installing more sensors, actuators and intelligence in a building is just figuring out how the building is put together. To an extent this problem is less important during new construction, but it is evident that the majority of environmental benefits to be gained from smarter buildings will come from retrofitting the existing building stock.

Complex Efficiency Incentives
Efficiency-driven investments for smart buildings are typically made to reduce the perceived energy consumption of site appliances; for example replacing incandescent light bulbs with LEDs, and installing lighting controls. The source-energy savings from site-based efficiency improvements are difficult to forecast or measure because the systemic energy analysis of multiple site-appliances requires in depth system modeling. For example, turning off a few lights in one building does not translate to identical reduction in power output from the local power station. Likewise, an incremental reduction of the cooling load in a building does not yield the same incremental decrease in power consumption at the chiller that provides chilled water for the building's air-conditioning. The complication in both of these situations is due to the non-linear response of source-energy consuming appliances (such as a natural gas fired chiller or coal-fired power plant) to changes in the loads that are placed on them. Furthermore, the details of that response are also typically a function of how multiple loads are connected to a source-energy consuming appliance; again without as-built drawings to describe those connections, we cannot easily predict how an individual smart building will ultimately reduce its source-energy consumption and carbon emissions.

Depending on how a smart building attempts to reduce its utility costs, building appliances in a smart building may also be forced to operate in an extremely in-efficient manner even though they reduce the energy costs for the building. For example, the use of a grossly over-sized cogeneration unit to drive down electricity loads during peak demand hours of the day is a troubling practice. Cogeneration can in fact generate electricity at much higher efficiency than a typical central power plant, but only when the heat from that cogeneration unit is captured and utilized by the building. When operating an over-sized cogeneration unit to drive down peak electricity consumption, far too much heat is produced by the unit than can ever be used by the building and hence much of that heat is rejected to the outside air. The net result is reduced monthly electricity bills at the cost of extraneous natural gas combustion and carbon emissions.

Turning from appliances to tenant-landlord and tenant-tenant relationships, we find typical split incentives between those who inhabit spaces and those who pay the utility bills. For management agencies which charge a fee proportional to building operation costs, higher utility bills represent a profit center. Likewise sub-metering among different tenants in a building can be stonewalled by the tension between tenants who feel they are paying too much and those who enjoy paying too little. Overall, regardless of the perceived energy or environmental impact of advanced building controls and automation, devising the right incentive for efficiency is not a trivial effort.

Prevalence of Custom Rather Than Standardized Design/Build Construction
Standardization is a major driver for the success of quality control and automation in the manufacturing industry. Every manufacturing line is tooled specifically for a product that is supposed to be made the same way by the same machine, every day of operation. Buildings, on the other hand, are nearly all custom-built and designed; almost no two buildings are exactly alike. Different contractors can have different preferences for sensor placement, equipment specification, control logic, piping, wiring, ducting, finishing or any other aspect of construction. Consequently different buildings could have altogether different responses to the same exogenous weather or occupancy conditions, despite similar location, age, construction techniques or owner. This inherent variability in buildings is something that we cherish and celebrate through architecture, but it presents a significant challenge to the scaling of any smart building software or hardware technology across the building stock, or even across sectors of the building stock.

Technical Training and Risk Aversion in the Industry
Assuming that we had building intelligence technologies that were low cost, easy to deploy, and proven in test cases to work, how do we get the entire building industry to actually deploy them? This industry is inherently risk averse due to the scale of liability in construction, and technicians can be found with a wide range of training and expertise; even a silver bullet technology solves no problem if no one uses it.

For example, how do we convince contractors, who can be held liable for their installation problems, to install more advanced automation and control in buildings when it is hard enough to get existing complex systems to work the right way? The simple answer is when owners pay a premium for them to take on a higher risk job, or when contractors volunteer for it in order to expand their knowledge base. Volunteering only happens when contractors believes there will be more business for them after mastering that risk. In both cases, however, the choice is ultimately based on whether the contractor thinks that the new technology is really necessary and there is a market demand worth pursuing.

Final Thoughts
Despite these significant challenges, smart buildings continue to hold great potential for our built environment because we need to make buildings smarter in order to preserve our planet and way of life. The challenges outlined above are byproducts of an out-dated and shortsighted approach to real-estate development and building services. The future of building construction and maintenance rests within information technology, data-driven services and lifecycle decision making. Right now our building industry is experiencing the conflict between the out-dated methods of making buildings, and the incoming tools and techniques for raising an energy-efficient building stock. This transition period requires adoption of new perspectives on buildings, as well as new technologies for making and operating them. While the new technology can do much to improve building operations and energy efficiency, it is clear that technology on its own cannot fix buildings; only smart people working together in a concerted effort to change the built environment can fix up buildings for a 21st century world.

Stephen Samouhos is a Ph.D candidate in Mechanical Engineering at MIT, working in the Building Technology Laboratory on HVAC system fault detection and diagnostics. He also works part-time for his family's mechanical contracting company in West Orange, New Jersey.
Pass it on  

March 2017
Sun Mon Tue Wed Thu Fri Sat

Stay updated via Twitter, Google + or join our list.

Fulton Center transit hub oculus

We have made Snapshot articles available in .pdf format on Scribd.