28 Seconds: The Hidden Measure of Vertical Architecture

In office buildings, 28 seconds represent far more than a waiting time. A SiiLA analysis shows that the quality of vertical transportation depends more on the intelligence of the system’s operation than on the number of elevators installed.

Architect Richard Rogers co-designed Torre BBVA, which features 41 passenger elevators. Photo: SiiLA.

By: SiiLA News

08/07/2026

Ten elevators may be far too many for one building and completely insufficient for another. That apparent contradiction explains why simply counting elevators tells us much less about a building than we tend to believe. What matters is not how many it has, but whether its vertical transportation infrastructure matches the size and height it is designed to serve.

According to SiiLA data, Class A+ office buildings average eight elevators, Class A buildings four, and Class B buildings three. At first glance, the difference would seem to confirm that more elevators mean a better building. However, when gross leasable area (GLA) and the number of floors—which together approximate potential vertical transportation demand—are considered simultaneously, that difference virtually disappears, confirming that the number of elevators responds primarily to a building’s size rather than its commercial classification.¹

But a building’s dimensions only determine how many elevators it needs—not how they should operate. That is where system intelligence comes into play. Numerous studies show that adding elevator cars is often the least efficient and, in many cases, the most expensive solution. Instead, strategies such as elevator zoning, demand anticipation through sensors, and adaptive dispatch algorithms can reduce waiting times by about 20% on average and by as much as 70% under heavy demand, without adding a single additional elevator.²

The scale of the challenge can be seen in some of the country’s largest office buildings. In Mexico City, corporate towers such as Torre BBVA on Paseo de la Reforma have more than 40 elevators, while complexes such as Arcos Bosques in Santa Fe operate with roughly ten elevators per tower. In both cases, the challenge is not simply moving a car but coordinating dozens of simultaneous trips for thousands of occupants throughout the day.

So how can we tell whether an elevator system is truly performing well? The technical literature proposes two measures: service quality, reflected in how long users wait, and service capacity, determined by the system’s ability to move people during peak demand periods. As a benchmark, an office building should maintain average waiting times below 28 seconds and be capable of transporting between 12% and 15% of its occupants within five minutes during peak periods.³

However, vertical transportation demand does not remain constant throughout the day. Morning arrivals, lunchtime traffic, and evening departures generate distinct traffic patterns. Paradoxically, the highest volume of trips occurs during the lunch period, when the system must simultaneously serve people leaving and returning to the building. Moreover, peak periods are not continuous flows but a series of short-lived sub-peaks that hourly averages often conceal. As a result, an efficient system does not operate under a fixed strategy throughout the day but continuously adapts elevator assignments as demand patterns change.⁴

Seconds are also part of vertical architecture. Photo: BigStock.

Why does all this matter? Because, ultimately, elevators do more than manage the flow of people—they manage the time each occupant gains or loses inside a building.

In that regard, an experiment conducted in an office building found that users’ stress is more closely related to how long they believe they have waited than to the actual waiting time. Based on the model developed, the authors estimated that perceived stress increases significantly once waiting times approach one minute. The study also found that waiting generates more stress than the ride itself and that greater elevator occupancy further increases perceived tension. Consequently, improving an elevator system may also involve reducing uncertainty through real-time information, countdown indicators, or even simple features such as mirrors.⁵

Waiting times not only affect people. They also have economic consequences for buildings. A less efficient vertical transportation system results in higher operating and maintenance costs, lower occupant productivity and, ultimately, a reduced ability of the property to generate value. Part of that impact is explained by the fact that elevators typically account for between 2% and 10% of a building’s energy consumption, rising to as much as 40% during peak demand periods.⁶

Taken together, these findings show that an elevator is far more than a means of transportation within a building. It is infrastructure that simultaneously shapes occupant experience, operational efficiency and the property’s economic performance.

To learn more about Mexico’s office market, visit SiiLA Market Analytics or contact us at contacto@siila.com.mx.

***

¹ Results obtained using a Poisson generalized linear model with a log link function, estimated from a sample of 157 office buildings across the Mexico City, Guadalajara, Monterrey and Querétaro markets, with complete information on gross leasable area (GLA), number of floors, year of completion and number of elevators. The model was justified by the absence of relevant overdispersion (dispersion index = 0.91). The specification included the natural logarithm of GLA, the number of floors, the year of completion centered at 2000 and the building’s commercial class (Class B as the reference category). GLA (p < 0.001) and the number of floors (p < 0.001) showed statistically significant associations with the expected number of elevators, while commercial classification did not produce statistically significant coefficients (Class A: p = 0.326; Class A+: p = 0.300). Standard errors correspond to the HC3 robust estimator. Additionally, 14 potentially influential observations were identified using Cook’s distance (threshold = 4/n). Re-estimating the model after excluding those observations preserved the main conclusions, supporting the robustness of the results.

² Liu, Dynamic Planning of Office Building Elevator Scheduling (2024); Choi, Cho and Bahn, An Energy-Efficient Elevator Operating System that Considers Sensor Information and Electricity Price Changes in Smart Green Buildings (2019); Pepyne and Cassandras, Optimal Dispatching Control for Elevator Systems During Uppeak Traffic (1997). According to these studies, zoning assigns specific floor groups to different elevators; demand anticipation uses sensors to detect a user’s intent before the call button is pressed; and adaptive dispatch algorithms continuously adjust elevator assignments based on system conditions and passenger flow to reduce waiting times and, in some cases, energy consumption.

³ Russell, Elevator Systems (2021).

⁴ Shi, Xu y Choi, Gaussian Analysis of the Elevator Traffic under the Typical Office Building (2024). Based on monitored data collected between 6:30 a.m. and 6:30 p.m., the authors identify three main demand periods (arrival, lunch and departure), finding that the highest traffic volume occurs during the lunch period (136 passengers), exceeding both the morning arrival (113) and evening departure (108) because outgoing and returning flows overlap. They also show that each period consists of several short-lived sub-peaks and conclude that system operation should continuously adapt to the prevailing demand pattern. The LS-SVM model achieved a substantially better fit for inbound traffic (R² = 0.9786) than for aggregate daily traffic (R² = 0.6345), demonstrating that different demand patterns exhibit different levels of predictability.

⁵ Shiomi, Kakio y Miyashita, How Long Is Too Long? Examining Waiting Times and Stress in Human-Elevator Interaction (2025).

⁶ Al-Kodmany, Tall Buildings and Elevators: A Review of Recent Technological Advances (2024).

Latam

Mexico

National

Office

Market Analytics

Market Trends

ABOUT SiiLA

Founded in 2015, SiiLA is the industry leading REsource for comprehensive commercial real estate market insights, news and events across Latin America. The SiiLA suite of innovative products drive greater accuracy, efficiency, and strategic advantages for top players in the commercial real estate industry.