Abstract: Accurate short-term prediction of indoor environmental parameters is important for proactive ventilation control and environmental risk warning in cage-free laying hen production. Multi-tier perching layer houses provide hens with perches, nests and multi-level activity spaces, but their complex spatial structure and airflow organization may lead to local environmental differences and short-term fluctuations. This study aimed to develop a short-term prediction method for indoor temperature, relative humidity and ammonia concentration in a multi-tier perching layer house based on multi-point monitoring data. Continuous environmental data were collected from one experimental multi-tier perching layer house. The indoor monitoring points were arranged in different perching areas, while one outdoor monitoring point was used to represent boundary environmental conditions. After data cleaning, time alignment and resampling at a 20 min interval, point-wise mean values and spatial ranges were calculated to describe environmental differences among monitoring points. Pearson correlation analysis was used to examine the relationship among temperature, relative humidity and ammonia concentration. A short-term prediction framework based on gradient boosting regression was then developed. For temperature and relative humidity, historical observations, outdoor environmental information and indoor–outdoor difference features were used as model inputs. For ammonia concentration, ammonia historical features were combined with temperature and relative humidity features to construct a multivariable prediction model. The most recent 7 d data were used as an independent validation set. In addition, recursive prediction was performed to evaluate the model performance for a future 24 h horizon, and 0.05 and 0.95 quantile regression models were established to generate 90% prediction intervals. The results showed that indoor environmental parameters differed among monitoring points. The mean values of temperature, relative humidity and ammonia concentration at different indoor monitoring points were 18.99～20.88 ℃, 46.52%～51.98% and 1.32～1.86 mg/m³, respectively. The corresponding mean spatial ranges were 2.31 ℃, 11.84 percentage points and 1.27 mg/m³, indicating that the monitored parameters varied among different perching areas. Correlation analysis showed that temperature and relative humidity were negatively correlated with ammonia concentration, with correlation coefficients of −0.26 and −0.16, respectively. This result suggested that ammonia concentration was affected not only by its historical state, but also by thermal and humidity conditions and ventilation-related processes. In the independent validation set, the temperature prediction model achieved a coefficient of determination of 0.96, a root mean square error of 0.73 ℃ and a mean absolute error of 0.50 ℃. The corresponding values for relative humidity were 0.97, 2.94 percentage points and 1.77 percentage points, respectively. For ammonia concentration, the multivariable prediction model achieved a coefficient of determination of 0.76, a root mean square error of 0.44 mg/m³ and a mean absolute error of 0.27 mg/m³. Compared with the univariate gradient boosting regression model using only ammonia historical information, the proposed multivariable model increased the coefficient of determination from 0.64 to 0.76 and reduced the mean absolute error from 0.37 to 0.26 mg/m³, showing that temperature and relative humidity features provided useful supplementary information for ammonia prediction. The 24 h recursive prediction results showed that the developed models maintained stable prediction performance during continuous forecasting and produced smooth prediction curves without obvious abnormal jumps. The models also provided 90% prediction intervals, which quantified the possible fluctuation range of environmental parameters. The proposed short-term prediction method can support proactive ventilation adjustment, ammonia risk identification and precision environmental management in multi-tier perching layer houses.