The Job Shop-Scheduling Problem (JSP), which considers the constraints of machines with aging effects and workpieces with release time, is studied. A scheduling optimization model with the objective of minimizing the maximum completion time is developed and an improved Arithmetic Optimization Algorithm (IAOA) is designed to solve the problem. The algorithm maps the IAOA continuous solution space to the discrete space of the JSP by means of ranked-order value (ROV) transformation rules, encodes the JSP and decodes it using an insertion greedy decoding algorithm. A non-linear mathematical optimization acceleration (MOA) function and six neighborhood search strategies are proposed to improve the standard Arithmetic Optimization Algorithm (AOA). The IAOA is compared with AOA, Grey Wolf Optimizer (GWO) and Arithmetic Trigonometric Optimization Algorithm (ATOA) by solving 33 benchmark problems. The experimental results show that the IAOA proposed has better optimization effect and convergence ability on JSP. The IAOA algorithm proposed overcomes the shortcomings of the AOA algorithm in terms of low solution accuracy and slow convergence speed.