This study aims to further elucidate the impact of age, period, and cohort on the fluctuations in forest workers population in proportion to a population. The age-period-cohort analysis was applied to the data on agriculture, forestry, and fishery workers from the national census from 1980 to 2015, and the influence of each factor was determined. The results indicate that the age strongly influenced and period weakly influenced the proportions. The most prominent worker proportion factor was the difference in age. At ≥75 years old, after 2000 years, this factor was the cohort. However, under 35 years of age, this factor notably differed in agriculture, forestry, and fishery industries. This is because the cohort effect of the individuals born after 1971 differs in the three industries. Moreover, the individuals born after 1971 seemed to engage in forestry more than those born before it. This is possibly due to the implementation of the employment support policy for the younger generation. Furthermore, no gender difference was observed in the cohort effect of the generation born after 1971. Therefore, the employment support policies were effective regardless of sex. Hence, these results will help design and implement efficient forestry employment support policies in future.

In recent years, conducting RTK (real-time kinematic) measurements in conditions such as the forest environment has become possible.This study evaluated positional errors of RTK measurements in terms of precision, accuracy, and 2DRMS in a sugi (Cryptomeria japonica) plantation forest with a low-cost dual-frequency GNSS receiver and high-precision positioning service (SoftBank Corporation). It was found that errors in precision (0.17–0.47 m), accuracy (0.85–0.93 m), and 2DRMS (1.78–2.24 m) were much greater than those under the open skies. Area measurements showed that the area calculated by the RTK coordinates was found to be −3.7%, which was satisfactory considering the small area measured. However, standing trees negatively impacted fixed solution rates, ranging between 15.4% and 29.9%. The t test found that there was no significant difference in the positional accuracy between fixed and float solutions. Therefore, float solutions which are equivalent to fixed solutions in the forest environment should be used instead. Although additional trials were conducted to improve the positional accuracy with a different type and higher height of GNSS antenna, no significant improvement was found.

Vehicle location data in the forest can be used in various areas, such as the automation of wood transportation. However, when using a global navigation satellite system (GNSS), it becomes difficult to obtain accurate real-time location information in the forest. Therefore, it is necessary to acquire vehicle location data using methods other than GNSS. Here, we applied our independently developed light detection and ranging (LiDAR) simultaneous localization and mapping (SLAM) system to forestry machines and investigated its accuracy. LiDAR SLAM for autonomous forwarders integrates inertial measurement unit and wheel encoder information. Additionally, the LiDAR SLAM matching process is divided into two stages: coarse and fine precision. Thus, a maximum and average error of 0.064 and 0.020 m, respectively, were obtained when the autonomous forwarder estimated its position in the horizontal direction. Additionally, a maximum and average error of 0.34 and 0.17 m were obtained during SLAM automatic operation. Therefore, it was confirmed that LiDAR SLAM could be applied as a self-position estimation method in a forest environment where it is difficult to use GNSS.