Abstarct: Vacuum ovens and furnaces have many applications in the industry. The most widely used and common type is the box type. Creating a homogeneous thermal environment in ovens and vacuum furnaces is of great importance. Also, accurate temperature control is very important to achieve the desired goals. To achieve these goals, having a mathematical model will be very useful. On the other hand, due to the presence of radiation heat transfer mode in a vacuum environment, the behavior of the system will be very nonlinear and therefore linear models do not have the necessary efficiency and accuracy. In this research, first, a new model of ordinary nonlinear differential equations for a vacuum box oven is extracted and the validity of the model is investigated using an experimental set. Then, by obtaining the system transfer function at three different operating points, its nonlinearity is investigated. Next, an ARX and ARMAX model for the system is obtained and then validated. To control the oven temperature, a PID controller is designed baxsed on the proposed model, the coefficients of which are obtained using genetic algorithms, invasive weed, firefly, and imperialist competition. In the next step, a fuzzy controller is designed in which the rules, input and output membership functions, and input and output gains are optimized using genetic algorithms, harmony search, cuckoo optimization, and the water cycle. Also, the measurement error in all optimization algorithms is ISE, IAE, ITSE, and ITAE. Comparing the results of the PID controller and the fuzzy controller, it can be seen that the result obtained in the fuzzy controller optimized by the water cycle algorithm has a better performance than the others. Finally, the PID and the fuzzy controller are implemented on the oven. The results show that the results obtained from the experimental tests are similar to the simulations performed.