ARCHIVED - "Anywhere anytime computing" for the construction industry
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.
December 07, 2007— Ottawa, Ontario
|MFDE improves the usability of mobile devices for construction workers.|
Tomorrow's mobile handheld PCs could give workers the ability to input data "anywhere anytime" – even without looking at or touching the device. That's the ultimate goal of an NRC research team that has created a flexible voice and pen-based data entry tool to improve the usability of mobile devices for construction workers.
"In construction, it's important to have real-time data from the field for timely decision making," says Dr. Irina Kondratova, who leads the People-Centred Technologies Group at the NRC Institute for Information Technology (NRC-IIT) in Fredericton. "People usually record information using pen and paper and then go back to their offices to enter it into a computer, which is inefficient."
She notes that there are many other work environments where people need to collect data outside an office, such as large-scale agriculture, manufacturing, emergency services, health care, and public safety inspections. "Again, it's the same scenario: people are working in the field," says Dr. Kondratova. "Their hands are busy and their eyes might also be busy because they're watching or monitoring something. In these settings, speech would seem to be a natural way of interaction with mobile technology."
In collaboration with civil engineering professors at the University of New Brunswick (UNB) – and through them cluster stakeholders such as the construction industry and the New Brunswick Department of Transportation – the NRC research team developed a prototype multimodal field data entry (MFDE) application to facilitate concrete inspections at construction sites. The MFDE system allows engineers to use either speech or a stylus to enter data from concrete tests into an electronic form.
"The idea is to give workers more than one means of accessing and entering data in their computing device, when they are using instruments or taking measurements," she says. "So even though they're busy doing other things, they can use speech to enter information at the same time."
She and her colleague, Dr. Joanna Lumsden, evaluated the prototype data-entry system using NRC-IIT's Mobile Human-Computer Interaction (HCI) laboratory. "Inside our HCI laboratory, we can simulate a real-life workplace context and conduct experiments in a controlled environment," says Dr. Kondratova. The research team evaluated the MFDE system with study participants recruited from different UNB faculties.
The NRC researchers used a surround sound system to simulate real construction noise recorded at an actual construction site. The MFDE system was then tested over three noise level ranges: 70-80 decibels, 80-90 decibels and 90-100 decibels.
|A University of New Brunswick student tests the MFDE system in NRC-IIT's Mobile Human-Computer Interaction (HCI) lab.|
"We also recreated common distractions that people might encounter on a construction site by projecting images on the laboratory wall. Some of them were 'safe' images while others depicted 'danger' scenarios – machinery that people need to watch out for," adds Dr. Kondratova. "We wanted to make sure our technology would not impair people's awareness of their surroundings and affect their safety."
In the lab trials, the subjects "really liked the capabilities provided by speech-data entry," she reports. "They performed tasks faster when using speech and they claimed to be more aware of their surroundings when using speech to enter data as opposed to a stylus."
Not surprisingly, the accuracy of speech recognition decreased at higher noise levels. "Yet, even though it was less accurate, participants still seemed to favour using speech for entering data," says Dr. Kondratova. As a result, the NRC team conducted another study to identify a better microphone for construction applications. Their research won "Best Paper" at the prestigious British Human-Computer Interaction conference, HCI 2007. "We also plan to look at different speech recognition engines to see if we can improve the accuracy at higher noise levels."
"Speech recognition technology that's built for an office environment might not be applicable to a mobile environment, which is much more complex," she notes. "There is still a lot of research needed in order to evaluate the technology, especially for different industries and workplace environments."
According to a study conducted for the research team by the NRC Canada Institute for Scientific and Technical Information, there are very few research groups exploring the use of speech-data entry in industrial settings – especially in construction. "So this creates an opportunity for our group to carve out a new niche. Speech data-entry for mobile devices is pretty novel, and evaluating the feasibility and usability of using speech in the field is also novel," she concludes.
- IT & Communications : NRC's Areas of Research
- NRC's Institute for Information Technology, People-Centred Technologies Program
Enquiries: Media relations
National Research Council of Canada
Report a problem or mistake on this page
- Date modified: