Research findings – Measuring well-being in real-time

The well-being of operators can be measured by studying arousal in real-time.

The objective of the studies was to show how the well-being of installers can be digitally measured, and demonstrate how data can be utilised and presented in real-time. Four technical solutions were tested.

Techniques 1–4 are presented below:

These solutions measure physiological data (heart rate, EEG, activity and temperature) during:

• 13 lab experiments in order to investigate how external factors (sound, light and temperature) affect the perception and performance of operators.

• Five user studies in which three activities were carried out in order to test the user-friendliness of the technologies.

In addition to physiological data, four work environmental factors are measured in real-time: temperature, carbon dioxide volume, light and the decibel level in the workplace. The field below the work environmental factors is for comments, where suggestions can be submitted to the user if thresholds are exceeded; for example, if the temperature becomes too high (above 23 degrees). A message that the level has been exceeded is issued, along with a proposed measure (not mandatory).

A prototype was developed in which data from the measurements is saved, in order to study physiological and work environmental data at a specific point in time.

The prototype was evaluated at a workshop, with experts invited to participate. The workshop consisted of 15 people (eight researchers, three company representatives and four project managers), and underlined that there are a lot of possibilities.

Strengths identified are that it is flexible and can be connected with a range of other technologies. It was also considered to be the first stage towards greater awareness of well-being in the workplace. The solution’s weaknesses included the difficulty of interpreting data (what is good, what is bad?) as well as personal privacy (who is to be granted access to data?).

Experiment results

13 experiments were performed in order to test the operator’s perception of physiological data. The objective was to investigate which technology was most relevant, and why (technique 4, the brain activity meter, is not included in the experiment). The experiment participants assembled eight lego modules, and were exposed to external disturbances (poor light, noise or elevated temperature) during the first four. It was performed at Chalmers Smart Industry Lab (CSI-lab), where an assembly station was constructed in partnership with CGM. The participants were:

• Evenly spread across three age groups (younger than 30, between 30 and 40 and over 40).
• 30 percent female, 70 percent male.
• Five novices, four intermediate level and four experts (at building the specific lego module).

Techniques 1 and 3 were considered to be most relevant, while techniques 2 and 3 were considered least relevant. This means there were some who thought technique 1 was most relevant who also thought technique 3 was least relevant, while those who thought technique 3 was most relevant also thought technique 2 was least relevant. As technique 3 was considered to be least relevant by some respondents, technique 1 was selected for implementation for the demonstrator. An interesting finding is whether or not a technology was relevant based on how the test subjects perceived themselves (that is, were aware of their body language). For example, one participant commented that she normally doesn’t sweat (is often cold) but was acutely aware that her heart rate began to beat more quickly during certain activities (as a result, she preferred technique 3). Some participants who preferred technique 1 said the other techniques could be of interest for long-term measurement, but that technique 1 seemed to be more detailed and relevant, provided you could understand the meaning of all the data.

For additional results, take a look at the scientific article here