Use of a priority lane to increase voluntary visits to a milking robot in dairy cows

Voluntary visits to the milking robot are the basis of automatic milking system functionality. Therefore, problems arise when cows are undermotivated to visit the robot. Cows with reduced competitive abilities, specifically those that are lame or low-ranking (or both), are at risk of lower visit frequencies. These cows typically have reduced autonomy as they must conform to the schedules of more dominant herd-mates. Solutions seeking to ameliorate these access issues may thus improve welfare and productivity. We evaluated the effects of providing lame or low-ranking cows with increased access to the milking robot via a priority lane, which provided an additional entrance to the robot for this group only. We aimed to understand the effects of the priority lane upon training duration, milking robot visit behavior, lying behavior, and hair cortisol. We further aimed to understand how losing and gaining access to the priority lane would affect milking robot visit behavior. We hypothesized that priority access would increase total milking robot visits and lying time, decrease milking interval variability and hair cortisol concentrations, and would not affect training duration. Twenty-four lame (mobility score ≥2) or low-ranking (lowest third of herd) cows were matched into 12 pairs of equal mobility, social ranking, or both. Each pair was split over treatments (i.e., priority lane and control). An additional 18 cows were included in the herd to increase competition but could not access the priority lane. Data were collected in 4 phases (pretreatment phase, training phase, treatment phase 1, and treatment phase 2) over a 15-wk period. New priority and control groups were created for treatment phase 2. Training time and hair cortisol concentrations were compared using linear regression, and generalized linear mixed models were used to analyze milking and lying behavior during treatment phase 1. Wilcoxon signed-rank tests were used to compare values between pre- and postregrouping in treatment phase 2. During the training phase and treatment phase 1, priority cows had more lying bouts, and successful, unsuccessful, and total robot visits. No treatment differences in training time, hair cortisol concentrations, lying time, or milking interval variability were observed. During treatment phase 2, cows granted access to the priority lane increased their successful milking visits, whereas cows removed from the priority lane did not decrease their visits. In conclusion, priority access had no effect on welfare indicators or training time. However, it did increase milking robot visit frequency. We suggest that the productivity and welfare benefits of a priority lane may be greater under conditions of greater competition.