Elevator Energy Optimisation Tips
December 5, 2022
V6.4.4 BETA
How we calculate your results
AdSimulo V6.4.1 Beta is now available. Explore the new features and share your feedback.
AdSimulo V6.4.1 Beta is now available. Explore the new features and share your feedback.
When a building opens with too few lifts, the cost shows up every weekday morning. Tenants queue in the lobby, the managing agent fields complaints, and retrofitting a new shaft into a finished core can run into seven figures. Lift traffic analysis exists to stop that outcome before a single beam is cast.
This guide answers: what is lift traffic analysis, the metrics it produces, the methods used to calculate them, and how modern software replaces a week of spreadsheet work with a single set of runs. If you design buildings, specify their services or manage them once they are occupied, this is the discipline that decides whether the vertical transportation works.
Lift traffic analysis is the process of predicting how passengers will move vertically through a building and testing whether a proposed lift installation can carry that demand to an acceptable standard of service. It answers three questions a developer cares about: how many lifts, how big, and how fast.
The analysis models a population of people arriving, requesting floors and being served by a group of lifts over a defined period, usually the busiest five minutes of the day. From that model it reports a small set of numbers that together describe quality of service.
The headline metrics are:
A design is judged against target values for these figures. A prestige office tower might aim for an interval under 25 seconds and an average waiting time under 20 seconds; a budget residential block can tolerate more.
Lift cores are fixed early and changed late. The number and size of shafts is locked in during concept design, long before anyone counts the actual occupants. Get the demand profile wrong and there are only bad options: accept poor service, sacrifice lettable floor area to a wider core, or pay for a structural alteration that nobody budgeted.
This is why the analysis happens before the design is frozen. By simulating the morning up-peak against the real population, you find out whether four lifts are genuinely enough, or whether the building needs five, while it is still cheap to add one on paper.
There are two established ways to produce the numbers.
The traditional method computes round trip time, the average time for one lift to complete a full cycle up and back, using the building’s floor count, population, car capacity and speed. From round trip time you derive interval and handling capacity with a set of equations refined over decades and codified in standards such as CIBSE Guide D. Calculation is fast and transparent, and for a simple up-peak in a standard office it remains a sound first pass.
Calculation assumes idealised conditions: passengers arrive at a steady rate and lifts behave predictably. Real buildings are messier. People arrive in clusters, traffic flows in two directions at once, and modern dispatch algorithms make decisions that no closed-form equation captures.
Simulation models each passenger and each car as individual agents over time. It generates thousands of randomised arrivals, runs the actual dispatch logic, and records what every passenger experiences. The result reflects how the building will really behave, including the awkward mixed-traffic periods that calculation glosses over. For destination-dispatch systems, sky lobbies and mixed-use buildings, simulation is the only method that gives a trustworthy answer.
Consider a 15-storey office with 1,200 occupants and four lifts. A calculation might show handling capacity of 11% and an interval of 30 seconds during the up-peak. Against a target of 12% and 26 seconds, the design falls short. The analysis tells you this in minutes, and lets you test the fix, a faster lift, a larger car, or a fifth shaft, before committing.
That is the entire value of the exercise: every change is tested in software, where it costs nothing, instead of on site, where it costs everything.
For years this analysis meant bespoke spreadsheets and desktop tools that took specialist knowledge to drive. AdSimulo was built to remove that barrier. You set the building’s tenancy type, floor count and population, and its Expert System runs thousands of simulations to return the lift configuration that meets your target level of service, automatically, rather than leaving you to iterate by hand.
The consultants who have switched describe the difference in productivity rather than features. One chairman of a lift consultancy put it plainly after moving across: he would not be going back to his old desktop tool because the newer approach made his team far more productive. That is the practical effect of automating the search for a working design instead of testing options one at a time.
The software also produces the evidence. Twenty-three report types cover waiting times, handling capacity and journey times in a format ready to send to a client, and a 3D visualisation lets you watch passengers and cars move through the building during the up-peak so a queue or a bunched pair of cars is obvious at a glance.
Lift traffic design in the UK is governed primarily by CIBSE Guide D, the reference that defines the calculation methods, recommended service criteria and population assumptions the industry works to. You can find the official scope of that guidance through the CIBSE publications library. Anyone producing traffic analysis professionally should treat Guide D as the baseline against which their figures are read.
If you are sizing a building, start by establishing the population and the tenancy type, because every metric flows from those two inputs. Then decide whether a calculation will do or whether the building’s complexity demands simulation. For anything with destination dispatch, multiple zones or mixed use, choose simulation.
For a deeper walk through the metrics and how AdSimulo produces them, see the full lift traffic analysis service page, which links to worked examples and the underlying science.
None. “Lift” is the British term and “elevator” the American one for the same equipment, so the two phrases describe an identical process. UK and European practice tends to say lift traffic analysis; North American practice says elevator traffic analysis.
At concept and scheme design, before the lift core is fixed. The whole point is to influence the number and size of shafts while changing them is still inexpensive. A second pass is common at detailed design to confirm the final specification.
Calculation suffices for a simple up-peak in a conventional building. Simulation is necessary whenever the building uses destination dispatch, has sky lobbies or mixed tenancy, or where two-way and interfloor traffic materially affect the result.
It depends on building grade. A high-quality office commonly targets an average waiting time below 20 to 25 seconds during the up-peak, while residential and budget commercial buildings accept longer. The target is set against the standard of service the client is paying for.
