Simulating thermal comfort of passengers

Thermal comfort of passengers in cars, planes, busses and trains is getting more and more important. On the one hand it is an important selling point for mid-class and upper-class cars. On the other hand it is generally a fundamental aspect for customer satisfaction in any mode of tranportation. While a thermally comfortable environment often is not noticed consciously at all, an uncomfortable one certainly is.

New ways of improving the comfort within passenger compartments are constantly being researched. Specialized materials and coatings are checked for their suitability in reducing the negative effect of extreme environmental conditions. Examining their impact on the thermal conditions within a passenger cabin by simulation drastically reduces the costs when compared to physical experiments. The time necessary to test different variants is considerably reduced as well.

Determining the thermal conditions within passenger compartments is one thing. The other one is to combine these with the thermophysical sensation of a human being to get feedback about the sensed comfort level. With the thermal manikin FIALA-FE we offer a widely acclaimed virtual human thermal model based on the doctoral thesis by Dr. Dusan Fiala. This model, supplemented with a selection of comfort models, allows detailed analysis of the comfort sensation of human passengers.

Rating thermal comfort of passengers

Physical activity level

Every human generates internal heat. In a simulation the amount of heat generated is lumped into a quantity called "activity level". The activity level can be thought of as an index of the intensitiy of work or action performed by the subject. Its unit is the MET, the Metabolic Equivalent of Task. It is normed such that the basal activity level in resting state lies between 0.8 and 1.0 MET. For heavy physical work and normal sports activities, values up to 10 met are reached. Within a car cabin simulation, typically one specifies a value of 1.2 MET for the driver and 1.0 MET for other passengers.

Schematic showing the Metabolic Equivalent Level for Typical Activities
Typical activities and their metabolic equivalent (MET)
Schematic showing Metabolic Equivalent Level of Reading
Schematic showing Metabolic Equivalent Level of Playing an Instrument
Schematic showing Metabolic Equivalent Level of Working on a Computer
Schematic showing Metabolic Equivalent Level of Gardening
Schematic showing Metabolic Equivalent Level of Biking
Schematic showing Metabolic Equivalent Level of Climbing Stairs
Schematic showing Metabolic Equivalent Level of Jumping Rope
Schematic showing Metabolic Equivalent Level of Running
Typical activities and their metabolic equivalent (MET)

Influence of clothing

Image of Multiple Layers of Clothing on top of each other

The human body together with its surroundings can be viewed as a thermal system. Depending on the worn clothing, a human body is insulated from the surroundings at different rates. Essential for the insulation effect are thin layers of air encased within the clothing layers. In contrast to solid materials such as wool or any other fabric, air is a very bad conductor of heat and thus leads to high insulation values. It has to be considered that this encased air is squeezed out in contact areas, e.g. when a person sits down onto a seat. This translates into reduced clothing insulation effect in contact areas such as the seat. On the other hand, the seat itself can be considered as a virtual additional clothing layer. In extreme cases the insulation of the contact area is so effective that no heat is dissipated from the human body at all, ultimately leading to sweating of the person sitting on the seat.
THESEUS‑FE allows its users to freely change the virtual human's clothing. Typical compositions representing a summer and winter outfit come shipped with the software and are ready-to-use.

Short-term and long-term thermal comfort

Image showing Contact Area of Seat and Manikin
Contact area between seat and body

Many of the comfort models used in practice are based on a quasi-stationary point of view. That means that the short-term comfort behavior is neglected. An example for this is the cooling effect sensed in the seat contact area for a short time after sitting down. In the long run, due to the high insulation discussed above, heat accumulates and the human body might start to sweat.

Thermal neutrality

A thermal system is said to be at neutral state when it is in thermal balance with its surroundings. For the human body this translates to the situation in which all internally generated metabolic heat is fully transferred to the surroundings and physiological regulatory mechanisms like sweating and shivering are turned off.

Schematic showing Typical Temperature Results and other Quantities for Thermal Neutrality State of Virtual Human Thermal Model
Exemplary results for thermal neutrality and associated human body temperature values

Indicators for estimating the comfort level

PMV index

The comfort index PMV stands for "predicted mean vote" and stems from real-world trials with test persons rating their comfort sensations when exposed to different thermal conditions. The method was developed by Fanger and the results are presented on a seven-point scale from cold (-3) to hot (+3). The point of thermal neutrality is set to be at a PMV value of zero. The temperature of a human body (or rather that on its surface) remains contant over time at PMV=0.

  • PMV < 0:
    cool conditions where more heat is transferred to the surroundings than is produced internally. The temperature of the human body decreases. At extreme conditions the thermophysiology of the human body reacts with shivering to increase the internal heat production.
  • PMV > 0:
    warm conditions where less heat is transferred to the surroundings than is produced internally. The temperature of the human body rises. Ultimately, the human body reacts with sweating as a regulatory cooling mechanism.

Details on this kind of comfort value can be found in numerous standards, for example ASHRAE 55 and DIN EN ISO 7730.

Global comfort

The simplest statistical approach is to view the situation PMV=0 as the optimal case. Every deviation from this state means more or less discomfort. Another index by Fanger called PPD (for "percentage of persons dissatisfied") can be used to analyze the average comfort level. It takes into account that a uniform comfort sensation for different people plainly does not exist and postulates that even at PMV=0 at least approx. 5% of the people will be dissatisfied with the thermal climate.
Another approach also available within THESEUS‑FE is to estimate the global comfort by using the mean skin temperature of the human. Two models are the TS (thermal sensation) and DTS (dynamic thermal sensation) indices.

Measured variables for judging the global comfort

Schematic demonstrating various Environmental Effects Acting on a Human Body and its Thermoregulatory Mechanism
Boundary conditions acting on human body including its own regulatory mechanisms

For judging the local comfort either the local skin temperature values (or their time derivatives) can be used. Alternatively, the heat flow transmitted from the skin to the surroundings can be used.
Within a simulation with THESEUS‑FE the heat flow between the virtual human and its surroundings is split into four parts:

  • convective heat flow (between the nearby air and the skin or clothing respectively)
  • heat exchange resulting from short-wave radiation, e.g. solar radiation
  • heat exchange resulting from long-wave radiation, i.e. heat exchanged between the virtual human and the car cabin
  • heat flow within contact areas

These heat flows are highly dependent on the surface temperature of the virtual human. For naked body parts (e.g. the hands) this is equal to the skin temperature. For clothed body parts it is the temperature of the clothing itself that is to be used.

Equivalent temperature

Another quantity often used when evaluating the thermal comfort of a person is the so-called "equivalent temperature". In plain terms it is the temperature that one "actually feels". One reason for introducing this quantity is that the heat flow derived from a simulation on its own is not very meaningful for most people. Most people will tend to judge the sensation of heat by some kind of familiar temperature value. The definition of the equivalent temperature is inspired by the following question: How high would the temperature within an ideal, enclosed room with homogenous air and wall temperature and without forced air convection need to be to produce the same heat flow at the considered body part? This temperature value is then defined as the equivalent temperature of the body part in question.

Image of Exemplary Thermal Comfort Results for a Sitting Human
Results of a local comfort model based on equivalent temperature on a five-point scale

Local comfort models

THESEUS_FE supports several models for evaluating the local comfort. A well-known example is the Zhang model from UC Berkeley which basically rates the local skin temperatures of different body parts. In contrast to that, the index from ISO 14505 uses local equivalent temperatures and distinguishes between summer and winter clothing. A local comfort model based on equivalent temperatures has the benefit that each local comfort index of any body part can easily be traced back to the heat flow at its surface. This considerably simplifies the process of understanding the reasons for comfort or discomfort in a given situation.

Related publications available as free download

The Application of Thermal Simulation Techniques for Seat Comfort Optimizations [english, 0.6 MB]
S. Paulke, E. Kreppold
P+Z Engineering, BMW Group


Simulation der Fahrzeugklimatisierung mit lokaler Komfortbewertung [german, 6.0 MB]
S. Paulke, M. Ellinger, S. Wagner
P+Z Engineering

Thermomanagement im Automobil, CTI Forum | February 12-13, 2008 | Bonn, Germany

Auslegung und Beurteilung des thermischen Raumklimas in Fahrzeugen mit der Software THESEUS-FE [german, 1.3 MB]
S. Paulke
P+Z Engineering

Haus der Technik Seminar | October 14, 2008 | Munich, Germany

Thermal Comfort Design and Assessment of Vehicle Cabins with THESEUS-FE [english, 1.2 MB]
S. Paulke
P+Z Engineering

Haus der Technik Seminar | October 14, 2008 | Munich, Germany

The Influence of the Glass Material on the Car Passengers Thermal Comfort [english, 2.1 MB]
C. Neacşu, M. Ivanescu, I. Tabacu
SC Auto Chassis International, University of Piteşti


The Human Thermal Comfort Evaluation inside the Passenger Compartment [english, 0.9 MB]
M. Ivanescu, C. Neacşu, S. Tabacu, I. Tabacu
University of Pitesti, SC Auto Chassis International


Numerical Simulation of Car Cockpit Heating during Winter [english, 0.8 MB]
C. Neacşu, M. Ivanescu, I. Tabacu
SC Automobile Dacia SA, University of Piteşti


Studies of the Thermal Comfort inside of the Passenger Compartment using the Numerical Simulation [english, 1.1 MB]
M. Ivanescu, C. Neacşu, I. Tabacu
University of Piteşti

Motors 2010 | October 7-9, 2010 | Kragujevac, Serbia

Comparative comfort simulations for winter load cases using global and local comfort models [english, 4.0 MB]
S. Paulke
P+Z Engineering

Haus der Technik, 3. Tagung 'Fahrzeugklimatisierung' | May 07-08, 2019 | Essen, Germany