One of the most harmful greenhouse insect pests is the Trialeurodes vaporariorum or most commonly known as the greenhouse whitefly. The easiest way to monitor the population of greenhouse whiteflies is by the use of yellow sticky paper traps. The insect count information from the traps can be used for analyzing insect behavior by constructing biological models. In this work, stochastic models describing the effects of temperature and the time of day on the flight behavior of greenhouse whiteflies were developed. Sticky paper images and temperature data were collected from an organic tomato seedling greenhouse by using integrated wireless imaging and environmental sensors. The greenhouse whitefly counts were determined by processing the images using an insect counting algorithm. From the results obtained, differences between the flight rates of the greenhouse whiteflies for different ranges of temperature were observed. The relationship was shown to be best fit using a double Weibull distribution function with an r² of 0.988 and mean squared error of prediction (MSEP) of 0.001. Using the model, it was found that the optimal temperature for flight of greenhouse whiteflies was around 20-26°C. From the real-time counting data, different daily peak flight times were discovered. The peak flight rates were modeled using multi-peak probability distribution functions where it shows that the multi-peak Gaussian distribution has the best fit with an r² of 0.961 and MSEP of 0.006. The developed models can be used for developing insect pest control methods such as fuzzy temperature control and pesticide application scheduling.