Why there are two adaptive thermal comfort model equations?

I found two different adaptive thermal comfort model equations from de Dear et al’s two papers.

  1. Comfort operative temperature = 18.9 + 0.255[mean outdoor daily effective temperature]*

… from the paper

de Dear, R., & Brager, G. S. (1998). Developing an adaptive model of thermal comfort and preference. ASHRAE Transactions, 104(1), 145-167.

  1. Tcomf = 0.31[Ta,out] + 17.8*

… where Tcomf means indoor comfortable temperature, and Ta,out means mean outdoor air temperature,

… from the paper

de Dear, R., & Brager, G. S. (2002). Thermal comfort in naturally ventilated buildings: revisions to ASHRAE Standard 55. Energy and Buildings, 34, 549-561.

I don’t have access to the latest ASHRAE Standard 55-2013 now, so I don’t know which one is actually adopted by ASHRAE.

So, can anybody know about adaptive thermal comfort model kindly advise?

Thank you!

Any advices, guys?

Much appreciated!


Sorry for not responding sooner. By default, the adaptive model used in Ladybug is always going to be the most-recent official ASHRAE standard (currently ASHRAE 55 2013), which is this one:

tComf = 0.31 * runningMean + 17.8

You can see this is true by looking in the Javascript source code of the CBE Comfort Tool (http://comfort.cbe.berkeley.edu/) and specifically at the formula in their ASHRAE comfort model (source code for the comfort models can be found here: http://comfort.cbe.berkeley.edu/static/js/comfortmodels.js)

The first formula that you posted there seems to be an older one that was probably released as the first-ever Adaptive standard and was later revised (since the first ASHRAE adaptive model was released in 1999/2000). It is likely that this older one was developed with fewer studies than the currently-adopted one. Still, it is worth noting that this first formula is fairly close to the correlation used in the European standard (EN-15251)

tComf = 0.33 * runningMean + 18.8

I am currently working in the ability to use the European standard in LB so that we have all parties represented.

So, in summary, it’s important to keep in mind that there’s quite a bit of “wiggle-room” around these correlations. As long as you get the general trend correct, you should be able to produce good passive designs You really only need a dead-accurate formula if you are showing compliance at the end of a design process.



I am letting you know that I just finished re-vamping the LB+HB adaptive comfort model and also finished off all of the features of a new adaptive comfort chart. You can get all of the new components by syncing with the github and I will try to get some tutorial videos up soon on how to use all of the new features that have been added.


Dear Chris, thank you very much for your advices which are as prompt and detailed as usual!

I’ve found the latest ASHRAE 55-2013 standards which has the T_comf 80% acceptability band defined as you explained.

Much obliged!