Background Fall juries are the most frequent childhood injuries. Prevention of falls requires first to predict how daily environments lead to children’s fall injuries.

Objective This study develops a simulation system to visualize the process of how young children aged 1 to 3 years old move, climb, and get fall injuries in home environments.

Methods First, using a unique device, the author conducted experiments to clarify the relation between children’s age (N=37, 13 to 49 months old) and their climbable height. To predict the risk of tipping over the furniture when children try to climb the furniture, the authors measured vertical and horizontal forces applied to the furniture using force sensors. In this paper, vertical indicates the direction parallel to the direction of gravity. Horizontal indicates the direction perpendicular to the direction of gravity The authors also measured the moving speed by crawling, turning over, and walking. Finally, using the game engine Unity, the author developed the simulation system using behavior characteristics based on the measured data.

Results The authors found the following equation: Climbable height H = 9.8 x months + 251.3 mm. The vertical force was 236% of children’s weight, and the horizontal force was 45% of children’s weight at most. The average moving speed by turning over, crawling, and walking was 0.2, 0.2, and 0.7 m/s, respectively. The developed simulator allows us to predict the process of fall injuries by just inputting a child’s age and the three-dimensional data of a daily environment. Such three-dimensional data can be measured easily by recent smartphones such as iPhones and Android phones.

Conclusions The authors clarified young children’s behavior characteristics related to fall injuries, such as climbable heigh as well as the moving speed by turning over, crawling, and walking. The authors also developed the simulation system to visualize the process of how small children aged 1 to 3 years old move and climb in home environments.

Acknowledgment: This paper is based on results obtained from a project commissioned by the Tokyo Metropolitan Government. This research is partially supported by the JST-Mirai Program from Japan Science and Technology Agency.