Summary for CIFE Seed Proposals for Academic Year 2015-16
Proposal number:
2015-04
Proposal title:
Space-Mate: Computational Modeling for Building and Occupant
Cooperative Sustainable Performance
Principal investigator(s) 1
and department(s):
Kincho H. Law
Research staff:
Renate Fruchter, Vladimir Bazjanac,
and Graduate Student: Flavia Grey
Total funds requested:
$ 65,000
Project URL for
continuation proposals
http://cife.stanford.edu/SeedProjects2015-16
Project objectives
addressed by proposal2
Usable, Buildable, Operable, Sustainable
Expected time horizon
2 to 5 years
Type of innovation
Breakthrough
Abstract
(up to 150 words)
1
2
Building performance simulation tools model and predict building
performance, but their accuracy is compromised by simplified simulated
occupancy. Occupant behavior is complex yet energy modeling software
represents it as deterministic and unchanging in hour-long periods of
time, which leads to discrepancies between model results and measured
performance. These discrepancies limit the use of the models both as a
predictive tool and real time post occupancy evaluation of the building.
We propose to (1) define a computational modeling framework for
building and occupant cooperative sustainable performance; (2) collect
correlated occupant and building performance data sets in real time (3)
develop a computational spatial-temporal-physiological occupant model
and a preliminary prototype Space-Mate. Real-time occupant state and
building performance data feeds will generate dynamic occupancy
information for building energy performance simulation and building
space adjustment to respond to the evolving occupant’s energy needs
and provide feedback to the occupant for potential sustainable behavior
changes.
The PI(s) must be academic council member(s) at Stanford.
For this and the next points, delete the answers that don’t apply to your proposal.
Engineering or Business Problem
We are designing with old, limited data!
Building performance simulation tools model and predict building energy performance, but their
accuracy is compromised by the approximations, assumptions and simplifications made in the
model. This is equally true for building performance simulation used to predict performance of the
future building and justify decisions considered in its design, as for real-time simulation used to
optimize building operation. In both cases, the simulation of internal loads is a critical determinant
of the results, and the basis for most of a building’s internal load is simulated occupancy.
Several research projects aim to improve the calculation of occupancy for use in building energy
performance simulation are already in progress (e.g. Annex 66). Most of these projects are
developing occupancy definitions based on statistics of measured and recorded occupant
presence in buildings and building spaces that are varied depending on building type and use.
Such methodology has a general shortcoming: It does not link the generated statistical information
with its cause – the reasons that determined the defined occupancy as such. In other words, such
statistics define records of what occupancy took place, with no indication why and of the involved
agents of change.
Occupant behavior is complex and stochastic yet current energy modeling software represents
occupant behavior as discrete, deterministic and unchanging in hour-long periods of time [1]. Such
occupancy definitions affect not only building use schedules, but also related variables such as
lighting, plug loads, humidity, and airflow, which leads to questionable results and a known gap
between these and the building’s actual performance. These discrepancies limit the use of the
models both as a predictive tool for building performance during design – directly impacting the
sustainability of the project - and as part of post occupancy evaluation of the building in the
operating and maintenance phase – limiting the analysis of the usability and operability of the
facility, which are three out of the four CIFE RFP goals.
When available, the building energy performance simulation help facility managers identify
discrepancies between actual energy consumption of the building and ideal levels, and can serve
as a basis for decision making regarding the operation of the building’s systems. Yet this
information is limited as well, for the output of these models only provides data regarding energy
consumption, and the facility manager can become blinded to other important goals for the building
and focus solely on reducing energy. However, beyond a building’s energy consumption, a central
corporate goal focuses on knowledge work productivity. In order to achieve this goal, it is critical to
focus on occupant well-being, which subsumes comfort as a health-related indoor building
performance metric in addition to satisfaction, collaboration, productivity and creativity, to name a
few. Consequently, energy consumption and occupant comfort and well-being need to be
addressed in an integrated approach. The knowledge productivity depends on the human
physiological state, the activity or task being performed, the location where the activity takes place,
and the technology infrastructure used to perform the activity. These spatial-temporal-physiological
occupant characteristics are hard to characterize due to their variable nature- in time windows from
second-by-second, minute-by-minute, hourly, daily, weekly, monthly, and seasonally- and thus do
not align with the simplified, deterministic assumptions of simulated occupant models used in
energy simulation tools.
Law
Space-Mate
2
Figure 1. Energy Flow and End Uses in the Jerry Yang and Akiko Yamazaki Environment and Energy
(Y2E2) building at Stanford University [2].
Figure 1 is an example that illustrates the importance of plug loads as the end-use of electricity in
buildings. Plug loads are largely determined by spatial-temporal-physiological occupant
characteristics. Determining these characteristics will help identify strategies for potential reduction
in the energy consumption of buildings such as Y2E2, without sacrificing occupant well-being.
What if the building space becomes a silent teammate to occupants’ activities as a continuous realtime dialogue between building energy performance and occupant state as they dynamically
interact and affect each other through co-simulation?
Current building sensor technologies and biometric sensors offer new opportunities to collect data
in real time from both the building’s operations as well as the occupant’s physiological state. Using
this data, we propose to (1) define a computational modeling framework for building and occupant
cooperative sustainable performance; (2) collect correlated occupant and building performance
data sets in real time; (3) develop a computational spatial-temporal-physiological occupant model
and a preliminary prototype called Space-Mate. Real-time occupant state and building
performance data feeds will generate dynamic occupancy information for building energy
performance simulation and building space adjustment to respond to the evolving occupant’s
energy needs as well as provide feedback to the occupant for potential sustainable behavior
changes. Moreover, the generated occupant models will make simulated occupancy in predictive
building energy performance simulation significantly more realistic than it is now.
Theoretical and Practical Points of Departure
This research builds on practical and theoretical points of departure that include: industry state of
the art such as the use of building energy performance simulations use -EnergyPlus, Ecotect,
eQuest- [3]; the use of building post occupancy evaluations to avoid, diminish and understand the
impact of occupant’s behavior [4], and the use of building automation systems [5]. We will leverage
the use of sensors and building sensor networks to obtain qualitative and quantitative data, thus
linking environmental sensors to physiological and health data. Towards the goal of developing a
more accurate occupant model and Space-Mate prototype we will explore and expand a number of
agent software models such as the open source ADAPT software (Agent Development and
Prototyping Testbed) [6] or the SAFEgrees (Social Agents for Egress) [7], and 3D ICC’s Immersive
Terf™ platform.
Although a variety of building energy and performance models exist in the literature, three general
types of modeling methods are widely used: physical models consisting of partial differential
equations of physical laws that govern energy flows in buildings, statistical models based on
experimental data, and the combination of both physical and statistical. Currently, the most
dominant method for modeling energy performance is the use of building energy performance
simulation tools such as EnergyPlus, Ecotect, and eQuest, which are physically-based.
Law
Space-Mate
3
Nevertheless, due to many assumptions and simplifications that are typically made during the
modeling process, predictions obtained from these simulation tools deviate from the actual
measurements, both in terms of energy consumption and indoor environment. These differences
have been attributed to three main causes: faults in the building envelope and systems, the
influence of occupants and weather changes. In particular, the accurate representation of
population heterogeneity presents a long-standing gap in energy and building modeling [8].
Post-occupancy evaluations (POE) have gained more importance in the last years, as an attempt
to better understand the role of occupant behavior and its impact on energy and building
performance. The have been defined as a “diagnostic tool and system which allows facility
managers to identify and evaluate critical aspects of building performance systematically”, and are
separated into three categories: indicative, investigative and diagnostic. Indicative POEs are based
on quick walkthroughs, structured interviews and inspections. Investigative POEs utilize interviews
and survey questionnaires, photographic/video recordings and physical measurements. Diagnostic
POEs require sophisticated data gathering and analysis techniques and traditionally take months
or years. None of these POE types include real-time data analysis and feedback, since they are
set up not as a continuous monitoring system but as one-off systems. For this reason,“interactive
adaptation” was introduced where an improved feedback loop with user information is established
for continuous information collection. [9]
Building automation systems have been established as part of the “smart buildings” trend with the
purpose of overcoming facility management challenges and realize the potential of improved
operational and energy efficiencies by optimization or automation of facility management work
processes and providing a wealth of data for analysis. Yet, despite advances in the building
automation industry, building management improvements have not kept pace. Some common
reasons for this, including a lack of understanding of facility design intent and inadequate training
have been listed in the literature [5] resulting in a struggle to operate the building as designed,
limited performance measurement and controls and a lack of standardization.
Research Methods and Work Plan
In the proposed framework we envision three layers with the respective units of analysis (Figure 2):
- The physical world represented by the building and the occupants
- The simulation world represented by building performance simulation software and agent
software
- The 3D immersive virtual world represented by the virtual building model and the avatar
agent model
We propose to launch the Space-Mate research effort. The CIFE seed project will serve as the
Stage one feasibility study. We will start with interviews and/or surveys to determine the current
state of practice and an extensive literature review to develop a holistic integrated perspective, and
a graphic summary to represent considered criteria and results to date. We plan to use big data,
data analytics and visualization techniques to cluster the criteria and develop a mechanism for
determining feature sets to be included in the Space-Mate model.
In Stage 1 of the project, a central objective will be to develop a protocol for concurrent building
spaces and occupant data collection. These two data sets will yield a longitudinal correlated space
and occupant database. Using this database, we will develop the spatial-temporal-physiological
occupant response probability distributions. We will use this dynamic occupant model to develop a
preliminary Space-Mate framework (Figure 2). Our target will be to identify two different types
spaces in the Jerry Yang and Akiko Yamazaki Environment and Energy (Y2E2) building at
Stanford University to be instrumented for data collection purposes; e.g. a room facing south
façade with continuously changing solar heat loads and occupancy variability, and an interior
Law
Space-Mate
4
space with no windows (e.g. CIFE or PBL lab). We will process the IRB protocol for Human Subject
Research in order to instrument the occupants of those spaces and collect physiological data.
Stage 2 will take the dynamic occupancy information generated during Stage 1, i.e. the correlated
building and spatial-temporal-physiological occupant data sets, for a preliminary building energy
performance simulation that is based on occupant response to the environment: occupancy
definitions resulting from occupants’ response to simulated change in environmental conditions
affecting individual occupants in the simulation (such as space temperature, exposure to solar
radiation, glare, comfort, air flow, etc.). The continuous recalculation of occupancy will require cosimulation. Environmental conditions will be calculated by the "new generation" version of the
EnergyPlus simulation engine. The "new generation" energy simulation engine will feed the
"current" environmental conditions data to another simulation tool: the agent software that will
calculate individual occupant response to those conditions (Figure 2).
Stage 3 will propose a new computational spatial-temporal-physiological building and occupant
performance prototype in a 3D immersive virtual world platform, with real-time input of
environmental and occupant data, which will provide feedback to the occupants about their state
as well as to the facility manager about the performance of the building in terms of well-being,
environmental impact and lifecycle cost (Figure 2).
The preliminary results will be documented, reported and disseminated and serve as preliminary
evidence to write a larger proposal for the continuation of the project and seek the support from
corporate and funding agencies.
Fig. 2 Space-Mate’s project stages
Expected Results: Findings, Contributions, and Impact on Practice
The expected findings and contributions of this research project focus on increasing the usability,
operability, and sustainability of building performance and occupant well-being.
Usable
- Correlated occupant and building performance data sets will allow for data analytics and
visualization studies to better understand occupant’s physiological state, how occupants
use and interact with and within the space.
- The occupant model developed throughout this project can be used to calibrate CIFE’s
Activity-Space-Performance (ASP) model [10].
Operable
- A computational spatial-temporal-physiological occupant model will expand current
occupant models that focus only on comfort metrics with spatial-temporal-physiological
occupant characteristics, which will provide new baseline targets for facility managers.
- Definition of appropriate time unit for real-time co-simulation of agent response to
environmental conditions, balancing micro and macro variations of occupant conditions and
heterogeneity of individuals.
Law
Space-Mate
5
-
Space-Mate will provide a framework for taking the facility manager’s role from simply
focusing on the building’s energy consumption and its related lifecycle costs, to a well-being
management position where the metric of building performance will be occupant well-being.
Sustainable
- Scalable database architecture/framework for a correlated building-occupant data sets
- For the Operation & Maintenance phase of the building life cycle: Real time dynamic
modeling and simulation environment to improve building sustainable energy performance
and well-being of occupants. The sending of continuously recalculated agent response to
environmental conditions back to the "new generation" energy simulation will update
occupancy in buildings and individual building spaces at the end of each time unit in the
simulation. This will make the simulated occupancy in predictive building energy
performance simulation significantly more realistic than it is now.
- For the Design phase of the building life cycle: This will provide more accurate occupant
models, building performance simulation engine, and correlated building-occupant dataset
framework.
Industry Involvement
We will interact directly with CIFE members by extending an invitation to participate in the SpaceMate project. Their participation will allow us to collect state-of-practice data and provide industry
specific input regarding needs and conditions of facility managers through interviews and surveys.
We invite industry to engage by providing data, new sensor technologies, and analytic tools.
Research Milestones and Risks
Fall
2015
Activity
Winter
2016
Spring
2016
Summer
2016
Literature Review
Interview/survey FMs
Develop data collection protocol
Process IRB for Human Subjects Research
Identify and instrument two spaces in Y2E2
Collect occupant data
Collect building data
Develop Space-Mate framework
Preliminary occupant statistics model for
agent
Rapid prototyping of simulation and agent
occupant information flow
Write larger proposal and seek funding
Document, report, disseminate preliminary
results
To collect statistically relevant data may require a longer time frame than the first year of this seed
project. To mitigate this risk we plan to continue the data collection effort over multiple years.
Law
Space-Mate
6
Balancing complexity and usability will be a research challenge. We plan to address this by rapid
prototyping and engaging industry partners and building occupants to provide input and feedback.
Next Steps
The proposed occupant and building performance data sets can serve as points of departure to
build further more accurate occupant agent software models and building performance simulation
software tools. This feasibility study will serve as a key point of departure to write an extensive
proposal and seek further funding.
Law
Space-Mate
7
References
[1] D’Oca, S., and Hong, T. (2015). Occupancy schedules learning process through a data mining
framework. Energy and Buildings. Pages 395-408.
[2] Graffy, K., Lidstone, J., Roberts, C., Sprague, B. G., Wayne, J., & Wolsk, A. (2008, March).
Y2E2: The Jerry Yang and Akiko Yamazaki Environment and Energy Building, Stanford University,
California. The Arup Journal, 44-55. http://www.arup.com/_assets/_download/DB23BA55-19BB316E-40F2B300E8544F3C.pdf
[3] Ham, Y., Golparvar-Fard, M. (2013, August). EPAR: Energy Performance Augmented Reality
models for identification of building energy performance deviations between actual measurements
and
simulation
results.
Energy
and
Buildings.
Pages
15-28.
http://www.sciencedirect.com/science/article/pii/S0378778813001485
[4] Guerra-Santin, O., Tweed, A. (2015, January). In-use monitoring of buildings: An overview and
classification
of
evaluation
methods.
Energy
and
Buildings.
Pages
176-189.
http://www.sciencedirect.com/science/article/pii/S0378778814008251
[5] Bhusari, S. (2014, June). Smart Building Integration. Consulting-Specifying Engineer.
[6] Shoulson, A., Marshak, N., Kapadia, M., Badler, N.I. (2013, November). ADAPT: The Agent
Development and Prototyping Testbed. Visualization and Computer Graphics. Pages 1035-1047
http://ieeexplore.ieee.org/xpl/articleDetails.jsp?reload=true&tp=&arnumber=6654163
[7] Chu, M., Parigi, P., Law, K., and Latombe, J.C. (2015, January). A Computational Framework
Incorporating Human and Social Behaviors for Occupant-centric Egress Simulation. CIFE
Technical Report #TR219. Stanford University. http://cife.stanford.edu/sites/default/files/TR219.pdf
[8] Lu, X., Lu, T., Kibert, C., and Viljanen, M. (2015, April). Modeling and forecasting energy
consumption for heterogeneous buildings using a physical–statistical approach. Applied Energy.
Pages 261-275. http://www.sciencedirect.com/science/article/pii/S0306261914012689
[9] Cole, R., Robinson J., Brown Z., and O’Shea M. (2008). Re-contextualizing the notion of
comfort. Build Res Information.
[10] Kim, T., Kavousian, A., Fischer, M. and Rajagopal, R. (2012, October). Improving Facility
Performance Prediction by Formalizing an Activity-Space-Performance Model. CIFE Technical
Report #TR210. Stanford University. http://cife.stanford.edu/sites/default/files/TR210.pdf
[11] Hasan, O., Defer, D., and Shahrour, I. (2014, October). A simplified building thermal model for
the optimization of energy consumption: Use of a random number generator. Energy and
Buildings. Pages 322-329. http://www.sciencedirect.com/science/article/pii/S037877881400560X
[12] Zhao, J., Lasternas, B., Poh Lam, K., Yun, R., and Loftness, V. (2014, October). Occupant
behavior and schedule modeling for building energy simulation through office appliance power
consumption data mining. http://www.sciencedirect.com/science/article/pii/S0378778814005714
[13] Maile, T., Fischer, M., Haymaker, J., and Bazjanac, V. (2010, August). Formalizing
Approximations, Assumptions, and Simplifications to Document Limitations in Building Energy
Performance
Simulation.
CIFE
Working
Paper
#WP126.
Stanford
University.
http://cife.stanford.edu/sites/default/files/WP126.pdf
[14] Lu, X., Lu, T., Kibert, C. and Viljanen, M. (2013, October). A novel dynamic modeling approach
for predicting building energy performance. Applied Energy.
[15] Dixon Smith, B., Kwok, A. Satterlee, R., Pincheira, F. and Howekamp, W. (2011, December).
Comparison of Y2E2 Occupancy, Comfort and Energy Audit to Building Objectives. CIFE Technical
Report #TR205. Stanford University. http://cife.stanford.edu/sites/default/files/TR205.pdf
[16] Kim, T. (2013, June). Predicting Space Utilization of Buildings through Integrated and
Automated Analysis of User Activities and Spaces. CIFE Technical Report #TR214. Stanford
University. http://cife.stanford.edu/sites/default/files/TR214.pdf
Law
Space-Mate
8
[17] Meier, A., Moezzi, M., Hammer, C., Goins, J. and Lutzenhiser, L. (2014, February). Behavioral
strategies to bridge the gap between potential and actual savings in commercial buildings.
California Air Resources Board.
[18] Gocer, O., Hua, Y., and Gocer, K. (2015, February). Completing the missing link in building
design process: Enhancing post-occupancy evaluation method for effective feedback for building
performance. Building and Environment. Pages 14-27.
[19] Deuble, M. and de Dear, R. (2012). Green occupants for green buildings: The missing link?
Building and Environment. Pages 21-27.
[20] Way, M. and Bordass, B. (2005). Making feedback and post-occupancy evaluation routine 2:
Soft landings – involving design and building teams in improving performance. Building Research
& Information. Pages 353-360.
[21] Wagner, A., Lutzkendorf, T., Voss, K., Spars, G., Maas, A., and Herkel S. (2014). Performance
analysis of commercial buildings – Results and experiences from the German demonstration
program ‘Energy Optimized Building (EnOB)’. Energy and Buildings. Pages 634-638.
[22] Menezes, A., Cripps, A., Bouchlaghem, D. and Buswell, R. (2012). Predicted vs. actual energy
performance of non-domestic buildings: Using post-occupancy evaluation data to reduce the
performance gap. Applied Energy. Pages 355-364.
Law
Space-Mate
9