Adaptive interactions between building occupants and their surrounding environments affect both energy use and environmental quality, as demonstrated by a large body of modeling research that quantifies the impacts of occupant behavior on building operations. Yet, available occupant field data are insufficient to explore the mechanisms that drive this interaction. This paper introduces data from a one year study of 24 U.S. office occupants that recorded a comprehensive set of possible exogenous and endogenous drivers of personal comfort and behavior over time. The longitudinal data collection protocol merges individual thermal comfort, preference, and behavior information from online daily surveys with datalogger readings of occupants’ local thermal environments and control states, yielding 2503 survey responses alongside tens of thousands of concurrent behavior and environment measurements. These data have been used to uncover links between the built environment, personal variables, and adaptive actions, and the data contribute to international research collaborations focused on understanding the human-building interaction.
behavior • humidity • Occupancy • degree of illumination • atmospheric wind speed • carbon dioxide • Demographics • temperature of air
Thermostat Device • Survey • sensor
Sample Characteristic - Organism
Sample Characteristic - Environment
Sample Characteristic - Location
United States of America
Machine-accessible metadata file describing the reported data: https://doi.org/10.6084/m9.figshare.9791435
Background & Summary
Humans interact with the built environment in a variety of ways that contribute to both building energy use and environmental quality and thus warrant significant attention in the building design, operation, and retrofit processes. Occupants’ thermally adaptive behaviors, for example – adjusting thermostats and clothing, opening and closing windows and doors, operating personal heating and cooling devices – are strongly tied to space heating and cooling loads, which in 2018 comprised 55% and 30% of total site energy consumed in residential and commercial buildings in the United States (U.S.), respectively1. These behaviors also modify key thermal comfort predictors like air temperature, air velocity and clothing insulation level2. In recent years, an occupant behavior research community has formed around a collection of studies that quantify the magnitude of occupant behavior’s influence on energy use and comfort3; the significant impacts reported in these studies have positioned behavior as a key topic of built environment research and related energy policy (e.g.4,5,6,7,8). In parallel, a new paradigm of occupant-centric building control has emerged that tunes centralized building operation strategies to real-time feedback on occupant presence, comfort and preferences9,10,11.
While the importance of the human-building interaction is well established, however, the mechanisms that drive this interaction remain largely undetermined. Improving the understanding of these mechanisms requires the collection of longitudinal data, which allow one to observe occupant comfort and adaptive behavior as they evolve together across the day and season12. Multiple longitudinal studies have been conducted in Europe across the past two decades, beginning with the Smart Controls and Thermal Comfort (SCATs) project13, which tracked thermal comfort, preference, and related behavioral adaptations from 1997–2000 in offices from twenty-five buildings across Europe and first developed models of the probability of a range of adaptive behaviors occurring over time. Subsequent studies have built upon the SCATS data collection approach, examining long-term comfort, window opening, occupancy, and environmental conditions in private and open offices in the United Kingdom14,15, Germany16, and Switzerland17, and using the collected data to develop regression-based models of occupant window interactions.
Such previous studies serve as important precedents for long-term data collection on human-building interactions; however, longitudinal measurements of building occupants are time-consuming and expensive to carry out, and publicly available comfort and behavior data remain limited in their coverage of certain adaptive actions, building types and climates. Few existing studies examine thermal behaviors in air-conditioned buildings, for example, or in buildings in climates with large seasonal variations. Moreover, existing studies rarely examine action hierarchies across several possible thermal behaviors, or the relationship between behavior diversity across an occupant population and individual-level thermal preferences. New longitudinal datasets that address such shortcomings are needed to improve the understanding of interactions between occupants and the built environment.
This paper introduces longitudinal data from a one-year study of occupant thermal comfort and several related behavioral adaptations in an air-conditioned office setting in the U.S. Offices were chosen as the context for the research because of their prevalence in the existing literature on occupant behavior data collection and because they are responsible for the most energy use of any single commercial building type in the U.S.18. The study develops a longitudinal protocol based in part on the data collection and analysis approaches of previous long-term occupant studies, adding new items that reflect the theoretical underpinnings of Humphrey’s adaptive principle19 and related theories of personal control from the psychology literature20. The primary objective of the data collection approach was to record a comprehensive range of exogenous and endogenous factors that may drive personal comfort and behavior outcomes over time, where the latter have not been explored in previous longitudinal studies of occupant behavior despite their suggested importance in previous behavior research21,22,23,24.
The longitudinal dataset includes a wide range of variables covering four of the six primary building performance measurement categories identified in the ontology developed by Mahdavi and Taheri25, as shown in Fig. 1. For the category of inhabitants, the data include information on occupant position in the building, available control actions, personal attributes and attitudes collected from online surveys. Datalogger measurements of indoor conditions span the hygro-thermal, visual and indoor air quality sub-categories, while concurrent measurements of outdoor conditions include the hygro-thermal sub-category only. Several occupant control actions were also measured on a longitudinal basis, using dataloggers where possible and online surveys otherwise; all data streams were merged based on time stamp as described further in the Methods and Data Records sections.
Longitudinal data on building occupant behavior, comfort, and environmental conditions were collected between July 2012 and August 2013 at the Friends Center office building in Center City Philadelphia, Pennsylvania, United States. Data collection proceeded in three stages, described further below: (1) semi-structured interviews, (2) site selection and subject recruitment for the longitudinal study, and (3) longitudinal survey and datalogger measurements. Drexel University Institutional Review Board (IRB) approval was received before collecting any of the data reported in this study (protocol #1204001220), and informed consent was obtained from all subjects.
Semi-structured interviews identify aspects of behavior that are not yet well known or understood and provide a rich qualitative context for developing and interpreting responses from structured survey instruments26. To inform the longitudinal portion of the current study, 32 interviews about thermal comfort and related behaviors were first conducted with office occupants from 7 air-conditioned buildings around the Philadelphia region. The building sample ranged from aging to recently renovated; small to large size; and suburban to urban. Interview responses were recorded and scored using the approach reported in Langevin et al.26.
Site selection and subject recruitment for the longitudinal study
The Friends Center office building in downtown Philadelphia was chosen for the full longitudinal study following its inclusion in the semi-structured interviews. At 5200 m2 and four floors (one sub-grade), the Friends Center represents a medium-sized office for the region. The brick-faced building, which was constructed in 1972, has a Building Automation System (BAS) following a renovation to LEED Platinum in 2009, and offers many adaptive opportunities to its occupants. Moreover, the building’s interior environment was reported to vary noticeably across the seasons and occasionally within a day, according to the occupant interviews.
Work spaces in the Friends Center include single and shared private offices (enclosed by full height, opaque walls with interior windows), partially open offices (cubicles defined by partial height, opaque partitions), and fully open offices; offices are located at either the core or perimeter of the building floor plan. Perimeter offices are each in close proximity to an exterior window, and the majority of perimeter offices face east, with smaller numbers facing north and west. Spectrally selective, high performance exterior glazing surfaces are offset by a few feet from the brick surface of the building facade, but are otherwise unshaded.
Subject recruitment was initiated through an e-mail message sent to all employees in the Friends Center by its Executive Director. The message outlined the general research approach, the project time period, and remuneration for participants. Included in the e-mail was a link to an online background survey that occupants could fill out to indicate their interest in participating in the full study. The survey was administered via SurveyGizmo and participation was compensated with a $5 gift card. The following question areas were included: (a) demographic information, (b) office characteristics, (c) thermal comfort and preferences, (d) control options, (e) personal values, and f) typical work schedule (arrival, lunch, departure times). A full background survey instrument is available for testing27.
The background survey generated a total of 45 occupant responses; from this initial sample, a final sample of 24 occupants was selected for participation in the full longitudinal study using a non-proportionate quota sampling strategy28. The strategy looked to achieve at least 1/3 of the final sample (or 8 occupants) in each of a series of key groupings, defined in order of importance by: gender; office type and location; behavioral control options; and comfort/perceived control satisfaction level. Table 1 presents the final occupant sample in terms of these categories.
Longitudinal survey measurements
Over the course of the following year, the final occupant sample participated in a series of subjective and objective measurements of thermal comfort, adaptive behavior, and related items. These measurements were carried via longitudinal online surveys, as well as through parallel datalogger and BAS measurements of the local environment and behavioral actions; the surveying approach is first outlined here.
Longitudinal surveys were distributed online via SurveyGizmo three times daily (morning, mid-day, afternoon) for a two week period in each of the four seasons, with the time of each survey distribution tailored to work schedules that occupants reported on the background survey. At the time of data collection, SurveyGizmo did not have a native function for repeated surveying of the same subjects; accordingly, the Gmail plugin Boomerang was used to send participants survey links at pre-specified daily times. To incentivize participation, occupants were offered a $5 gift card for every full day of surveying completed, in addition to the $5 previously received for participation in the background survey.
The daily survey instruments included questions about recent work flow, current thermal comfort and sensation, current and recent clothing levels, recent activity, and recent control use. Figure 2 diagrams the full progression of daily survey questions, indicating both the general question areas and associated areas of sub-questioning. A daily survey instrument is available for testing29.
Many of the daily survey questions shown in Fig. 2 follow a “show when” logic, where their appearance is conditional on a response to a previous question. The use of “show when” logic minimizes unnecessary questioning, keeping survey completion times down and combating possible survey fatigue. Fatigue was also countered through the use of several image response options, including a series of clothing checklist icons, for example.
Once the daily surveys had been filled out for the full two week period, each occupant completed a final survey that asked about the whole of their two week experience. A two week retrospective survey instrument is available for testing30. The final survey asked occupants to log any times when they were out of the office for a significant time (e.g., a whole or half day), also querying general comfort and the frequency of behaviors during the previous two week period. The survey also asked occupants to assess the accessibility, time expense, and general clarity of the daily survey instruments. Pairing this feedback with a close examination of the daily survey responses during the given two week period helped inform small revisions to the daily surveys between each seasonal run. Revisions included streamlining the general flow of questions and the addition of short follow-up prompts (for example, on the bodily location of reported thermal discomfort). Revised daily survey instruments were tested through two-day distributions in the time between the full two week seasonal runs to ensure the readiness of these revised instruments for the next two-week implementation.
Overall, the daily surveys generated consistently high response rates (>=85% in each season), yielding a total of 2503 responses from the 24 subjects. The surveys took a median time of just 2.3 minutes to complete and were positively evaluated by occupants. Table 2 summarizes key statistics regarding the performance of the daily survey instruments during each two week seasonal run as well as during the two day test runs in between.
Environment and behavior datalogger and building automation system (BAS) measurements
Alongside the longitudinal survey data collection, continuous measurements were made of the weather, the local indoor environment of each occupant’s office, and a subset of the occupants’ behavioral actions. Table 3 summarizes each of these measurements and their observed ranges during the two week surveying periods and across the entire year of the project.
Historical hourly weather data (temperature, humidity, and air velocity) were downloaded on a monthly basis from the Weather Analytics service (now Athenium Analytics) for the “Center_City_Philadelphia_580144” grid point.
Regarding interior conditions, ambient temperature was measured for every occupant using the nearest thermostat reading logged by the BAS system at 15 minute intervals. Ambient temperature and relative humidity were also logged at 5 minute intervals for 11 occupants using HOBO U12-013 base sensors (accuracy: +/−0.35 °C temperature, +/−2.5% humidity), as was globe temperature in eight cases (TMC6-HE attachment covered with 40 mm black matte ping pong ball; +/−0.25 °C accuracy), and air velocity in four cases (T-DCI-F900-S-O attachment; +/−0.05 m/s accuracy). Globe temperature sensors were placed on the perimeter and core areas of each floor space, and one air velocity sensor was placed on each floor. Additionally, illuminance and CO2 concentration were logged for certain occupants across floors using the HOBO U12-012 sensor (accuracy dependent on wavelength31) and the TelAir 7001 CO2 sensor (accuracy: greater of +/−50 ppm or 5% of reading), respectively. Sensor calibration was not performed after initial deployment. Further information on the location of each sensor including floor number, perimeter vs. core position within each floor, and office type are provided in the data record.
Datalogger measurements of occupant behavior recorded window adjustments and the use of personal fans and heaters. For windows, HOBO UX-90 state loggers were attached to 10 different windows that occupants in the sample reported using in the background survey. The UX-90 loggers record the time stamp of any state change during the measurement period. For fans and heaters, WattsUp? Pro power meters logged wattage at occupant-designated fan/heater outlets at 15 minute intervals. The WattsUp? meters were installed at the desks of all 9 occupants who reported using a heater or fan at some point throughout the year on the background survey.
Together with the daily survey responses, the above environment and behavior information comprises six longitudinal data streams (daily surveys, outdoor weather, BAS thermostats, local environment, personal fans and heaters, and windows). Figure 3 provides an overview of the collection and integration of these various data streams. The Data Records section provides additional detail on how each data stream was post-processed and matched with the other data streams.
A series of MATLAB32 scripts were used to process and integrate the environmental and behavior data streams listed in Table 3 as well as all survey data. The general post-processing and integration procedure used by these scripts is as follows:
Download the raw data at intervals specified in Table 3 (CSV format); survey data were downloaded at the end of the initial background survey period, and at the end of each two day test period and two week surveying period.
Remove any missing readings in the data or data points that are determined to represent faulty measurements based on observation of the building at the time of data download (e.g., measurement device was disturbed during measurement period or atypical operational patterns were observed during the measurement period).
Organize raw CSV data readings into larger CSV files along the following dimensions:
Measurement type (BAS, HOBO, weather, heater/fan, window, daily survey).
Occupant number (1 to 24).
Measurement variable (e.g., temperature, humidity, fan use, etc.).
Reading number (sequentially ordered from start to end of the study).
Use MATLAB scripts to merge all CSVs from step 3 into a single .txt file for each measurement type, mapping the CSV time stamps in date/time format to serial date numbers using the MATLAB datenum function.
Use a MATLAB script to merge .txt files from step 4 into a master .txt file that includes all measurement types in one place. In the script, disparate measurements are matched by time step in serial date form; where a certain measurement does not exist for a given time step (e.g., survey data in a time step where there were no surveys), a NaN value is shown. Additionally, a subroutine for calculating the Predicted Mean Vote (PMV) from measured environmental and occupant variables is used; this module follows the routine established in ASHRAE Standard 5533 for the PMV calculation.
The dataset that was produced by these steps is made available under an Creative Commons Attribution 4.0 International license through Zenodo34. All data are stored in a compressed .txt file (“LANGEVIN_DATA.txt”), which is accompanied by a data legend file in XLSX format (“langevincodebook.xlsx”). The data legend file characterizes each variable column found in the .txt file by type (environment, behavior, etc.), name, data collection source (surveys, dataloggers, etc.), value type (discrete, continuous), description, units, and response levels (for discrete variables).
Several measures were taken to ensure the validity of the collected data, following data collection guidance included in the final report for International Energy Agency Annex 66: Definition and Simulation of Occupant Behavior in Buildings35. These measures are outlined below.
Survey preparation phase
Longitudinal survey design was informed by preceding semi-structured interviews with office occupants along with background surveys of each occupant’s demographic characteristics, thermal comfort, and adaptive behavioral tendencies. Information collected during this preparation phase informed the design of longitudinal survey instruments, ensuring that the questions included on these surveys would be correctly interpreted by occupants and relevant to their daily experience of the office environment.
Encourage high response rates
The inclusion of financial incentives for participants and use of “show-when” logic to minimize survey fatigue yielded a high overall response rate of 86% across the longitudinal surveys (Table 2), ensuring the survey data are not biased towards respondents with higher participation rates.
Three small pilot studies were conducted during the course of the full longitudinal study: one in the month preceding the start of the study, during which time all required dataloggers were installed and collecting test data, and two in the periods between the summer and fall and fall and winter daily surveys, during which times improved survey deployments were tested alongside logger data collection. The pilot studies verified that all data collection devices and the online survey platform were functioning properly before each of the two week daily surveying periods was conducted.
Datalogger readings were downloaded regularly (at maximum every 2 weeks) to ensure their proper functionality and, in the case of environmental sensors, to verify that readings had not drifted significantly. Between these download times, sensor operation was visually inspected every few weekdays to ensure the devices were actively recording data and had not been moved or dislodged by occupants.
In the case of outdoor and indoor temperature, parallel data streams were compared to address internal validity. For outdoor temperature, hourly data points provided by Weather Analytics were spot-checked against concurrent measurements from the nearest weather station (KPHL) on Weather Underground. For indoor temperature, 5 minute interval HOBO temperature measurements were compared against concurrent BAS temperature measurements from the nearest thermostat. Here, summary statistics for whole year BAS temperature readings are within the margin of error of the HOBO temperature readings, indicating a close correspondence between the two measurements. Additionally, state logger measurements of behavior (heaters, fans, windows) were verified against corresponding daily survey responses that reported recent behavioral actions.
Comparison against expected conditions
One year inter-quartile ranges for each of the environmental variables in Table 3 are consistent with those expected of an air-conditioned U.S. office building with occupants that report high levels of general environmental satisfaction, specifically:
The range of ambient and globe temperature (22.0 °C to 24.0 °C and 22.0 °C to 23.9 °C, respectively) fall within the approximate range of acceptable operative temperatures in ASHRAE Standard 5533 (20.5 °C to 28 °C),
The range of relative humidity (25.7% to 52.9%) falls below the relative humidity threshold recommended by ASHRAE Standard 62.136 (at or below 65%),
The range of CO2 concentration (477 ppm to 637 ppm) falls below the indoor CO2 concentration threshold recommended by ASHRAE Standard 62.1 (below 1000 ppm), and
The illuminance range (28.0 lm/ft2 to 130.2 lm/ft2) encapsulates the average value recommended by the IES Lighting Handbook37 for commercial offices (40 lm/ft2), though the median illuminance level (75.4) is higher than the recommended value, likely reflecting the widespread access to daylighting in the studied building.
The longitudinal dataset described herein was first posted on OpenEI in July of 2015; in the years since, several researchers have contacted the author for clarifications on certain aspects of the data. Based on these inquiries, notes to guide further use and interpretation of the data are summarized below.
When reading in the main .txt file to a program such as R or Python for analysis, variables of interest should be indexed by the associated column number listed in the variable codebook file (“langevincodebook.xlsx”).
NaNs for certain variables do not generally imply missing data; rather, they indicate that the variable was not being measured for a given occupant at a given time stamp. For example, per Table 2 only four occupants had a direct illuminance measurement at their office location; for the occupants without this direct measurement, the illuminance variable will show NaNs across all time steps. Similarly, variables such as current and acceptable thermal sensation will show NaNs for all time steps that fall outside of the two week daily surveying windows.
The variable “Occupant Number” refers to an occupant ID (e.g., a value between 1 and the 24 occupants who participated in the study); it does not refer to an occupancy (presence/absence) measurement.
The variable “Occupancy 1” is an expected occupancy (presence/absence) value calculated across all time steps in the study based on the occupant’s responses about typical periods of occupancy on the background survey conducted before the start of the longitudinal measurements. The variable “Occupancy 2” is an expected occupancy state calculated across all time steps that fall under the two week daily surveying window, based on the time of arrival reported in the daily morning survey and recent departures from the office reported in the daily mid-day and afternoon surveys, as well as periods of prolonged absence reported on the retrospective surveys conducted after each two week period. Occupancy state was not continuously measured with dataloggers in this study.
Example applications of these longitudinal data in analysis and modeling efforts are found in both the author’s own work and those of other researchers pursuing research on occupant comfort and behavior. Regarding the author’s work, these data were used in38 to develop insights on the drivers of behavior (Fig. 4), long-term trends in thermal comfort and acceptability (Fig. 5), and the sequencing of adaptive actions (Fig. 6). The author also used the data to develop and validate an agent-based model of occupant behaviors in39, demonstrating the richness of the dataset for studies of either individual- or group-level behaviors in office settings.
Outside of the authors’ own work, several other researchers have used the longitudinal data to explore relationships between the built environment and occupant comfort and behavior, most notably: (1) the dataset was included in the recently developed ASHRAE Database II project40, a collection of 81,846 complete sets of objective indoor climatic observations with accompanying “right-here-right-now” subjective evaluations by the building occupants who were exposed to them, which is intended to support diverse inquiries about thermal comfort in field settings, and (2) the data were used in41 to develop synthetic profiles of building occupant personal characteristics and comfort preferences that can be used to intialize models of individual-level behaviors (for example, in an agent-based framework). The latter effort continues to be pursued and expanded through the recently approved International Energy Agency Annex 79 on Occupant-Centric Building Design and Operation42.
All MATLAB scripts used to implement the post-processing steps described in the Data Records section are available upon request, as are the raw data files described in steps 1 and 2 (e.g., pre- and post- cleaning). Interested readers should contact the corresponding author for access to the scripts so that further instructions on their use may be provided.
Additionally, links to the background survey instrument27, daily survey instrument29, and retrospective survey instrument conducted after each two week daily surveying period30 are available for testing.
U.S. Energy Information Administration (EIA). Annual Energy Outlook 2019, https://www.eia.gov/outlooks/aeo/ (2019).
Baker, N. & Standeven, M. A behavioural approach to thermal comfort assessment. International Journal of Solar Energy 19, 21–35 (1997).
International Energy Agency, Energy in Buildings and Communities Programme (IEA-EBC). IEA-EBC Annex 66: Defintion and Simulation of Occupant Behavior in Buildings, https://annex66.org/ (2019).
Delzendeh, E., Wu, S., Lee, A. & Zhou, Y. The impact of occupants’ behaviours on building energy analysis: A research review. Renewable and Sustainable Energy Reviews 80, 1061–1071 (2017).
D’Oca, S., Hong, T. & Langevin, J. The human dimensions of energy use in buildings: A review. Renewable and Sustainable Energy Reviews 8, 731–742 (2018).
Sun, K. & Hong, T. A framework for quantifying the impact of occupant behavior on energy savings of energy conservation measures. Energy and Buildings 146, 383–396 (2017).
Hong, T. & Lin, H.-W. Occupant Behavior: Impact on Energy Use of Private Offices. In ASim 2012 - 1st Asia conference of International Building Performance Simulation Association. (Shanghai, China, 2013).
Bourgeois, D., Reinhart, C. & Macdonald, I. Adding advanced behavioural models in whole building energy simulation: A study on the total energy impact of manual and automated lighting control. Energy and Buildings 38, 814–823 (2006).
Naylor, S., Gillott, M. & Lau, T. A review of occupant-centric building control strategies to reduce building energy use. Renewable and Sustainable Energy Reviews 96, 1–10 (2018).
Kim, J., Schiavon, S. & Brager, G. Personal comfort models – A new paradigm in thermal comfort for occupant-centric environmental control. Building and Environment 132, 114–124 (2018).
Nagy, Z., Yong, F. Y., Frei, M. & Schlueter, A. Occupant centered lighting control for comfort and energy efficient building operation. Energy and Buildings 94, 100–108 (2015).
Berquist, J., Ouf, M. M. & O’Brien, W. A method to conduct longitudinal studies on indoor environmental quality and perceived occupant comfort. Building and Environment 150, 88–98 (2019).
Humphreys, M. A., Nicol, J. F. & Raja, I. A. Field studies of indoor thermal comfort and the progress of the adaptive approach. Advances in Building Energy Research 1, 55–88 (2007).
Rijal, H. B. et al. Using results from field surveys to predict the effect of open windows on thermal comfort and energy use in buildings. Energy and Buildings 39, 823–836 (2007).
Yun, G. Y. & Steemers, K. Time-dependent occupant behaviour models of window control in summer. Building and Environment 43, 1471–1482 (2008).
Herkel, S., Knapp, U. & Pfafferott, J. Towards a model of user behaviour regarding the manual control of windows in office buildings. Building and Environment 43, 588–600 (2008).
Haldi, F. & Robinson, D. On the behaviour and adaptation of office occupants. Building and Environment 43, 2163–2177 (2008).
U.S. Energy Information Administration (EIA). Commercial Building Energy Consumption Survey (CBECS), Table E1. Major fuel consumption (Btu) by end use, 2012, https://www.eia.gov/consumption/commercial/data/2012/c&e/cfm/e1.php (2016).
Humphreys, M. A. & Nicol, J. F. Understanding the adaptive approach to thermal comfort. ASHRAE Transactions 104, 1–14 (1998).
Mansell, W. Control theory and psychopathology: An integrative approach. Psychology and Psychotherapy: Theory, Research and Practice 78, 141–78 (2005).
Kollmuss, A. & Agyeman, J. Mind the Gap: Why do people act environmentally and what are the barriers to pro-environmental behavior? Environmental Education Research 8, 239–260 (2002).
Clark, C. F., Kotchen, M. J. & Moore, M. R. Internal and external influences on pro-environmental behavior: Participation in a green electricity program. Journal of Environmental Psychology 23, 237–246 (2003).
Guagnano, G. A., Stern, P. C. & Dietz, T. Influences on Attitude-Behavior Relationships: A Natural Experiment with Curbside Recycling. Environment and Behavior 27, 699–718 (1995).
Hong, T., D’Oca, S., Turner, W. J. & Taylor-Lange, S. C. An ontology to represent energy-related occupant behavior in buildings. Part I: Introduction to the DNAs framework. Building and Environment 92, 764–777 (2015).
Mahdavi, A. & Taheri, M. An ontology for building monitoring. Journal of Building Performance Simulation 10, 499–508 (2017).
Langevin, J., Gurian, P. L. & Wen, J. Reducing energy consumption in low income public housing: Interviewing residents about energy behaviors. Applied Energy 102, 1358–1370 (2013).
Langevin, J. Daily Survey (Morning, April). SurveyGizmo, http://www.surveygizmo.com/collab/1204299/d0d5faf2bc2a (2012).
Fricker, R. D. Sampling Methods for Web and E-mail Surveys. In Fielding, N., Lee, R., Blank, G. & Fricker, R. D. (eds) The SAGE Handbook of Online Research Methods (Sage Publications, 2012).
Langevin, J. Background Survey. SurveyGizmo, http://www.surveygizmo.com/collab/872196/Background-Survey (2012).
Langevin, J. 2 Week Retrospective Survey (April). SurveyGizmo, http://www.surveygizmo.com/collab/1208532/5f365b9ef198 (2012).
Onset Corporation. HOBO U12 Temp/RH/Light/ External Data Logger (Part # U12-012), https://www.onsetcomp.com/files/manual_pdfs/13128-C%20U12-012%20Manual.pdf (2019).
MathWorks. MATLAB, https://www.mathworks.com/products/matlab.html (2019).
ASHRAE. ANSI/ASHRAE 55-2017, Thermal Environmental Conditions for Human Occupancy (ASHRAE, 2017).
Langevin, J. Longitudinal dataset of human-building interactions in U.S. offices. Zenodo, https://doi.org/10.5281/zenodo.3351509 (2019).
International Energy Agency, Energy in Buildings and Communities Programme (IEA-EBC). Annex 66: Definition and Simulation of Occupant Behavior in Buildings - Final Report (IEA-EBC, 2018).
ANSI/ASHRAE. ANSI/ASHRAE Standard 62.1-2016, Ventilation for Acceptable Indoor Air Quality (ASHRAE, 2016).
DiLaura, D. L., Houser, K. W., Mistirck, R. G. & Steffy, G. R. The Lighting Handbook: Reference and Application. (Illuminating Engineering Society (IES), 2011).
Langevin, J., Gurian, P. L. & Wen, J. Tracking the human-building interaction: A longitudinal field study of occupant behavior in air-conditioned offices. Journal of Environmental Psychology 42, 94–115 (2015).
Langevin, J., Wen, J. & Gurian, P. L. Simulating the human-building interaction: Development and validation of an agent-based model of office occupant behaviors. Building and Environment 88, 27–45 (2015).
Földváry Ličina, V. et al. Development of the ASHRAE Global Thermal Comfort Database II. Building and Environment 142, 502–512 (2018).
Senick, J. A., Andrews, C. J., Putra, H. C., Tsoulou, I. & Allacci, M. S. Synthetic populations of building office occupants and behaviors. In Preiser W., Hardy A. & Schramm, U. (eds) Building Performance Evaluation (Springer, 2017).
International Energy Agency, Energy in Buildings and Communities Programme (IEA-EBC). IEA-EBC Annex 79: Occupant Behaviour-Centric Building Design and Operation, http://annex79.iea-ebc.org/ (2019).
This work was supported by a U.S. National Science Foundation Graduate Research Fellowship. The author thanks Drs Jin Wen and Patrick L. Gurian of Drexel University for their invaluable guidance on the longitudinal data collection approach and implementation used in this study. The author also thanks Friends Center Executive Director Patricia McBee, Friends Center Building Manager Wil Mason, and all of the Friends Center employees for their gracious support of the longitudinal field study.
The author declares no competing interests.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Langevin, J. Longitudinal dataset of human-building interactions in U.S. offices. Sci Data 6, 288 (2019). https://doi.org/10.1038/s41597-019-0273-5
This article is cited by
Scientific Data (2022)
Scientific Data (2022)
Scientific Data (2020)
Scientific Data (2019)