<p>In a hierarchical Edge-Fog-Cloud computing environment, IoT devices have limited resources and power, and these devices can delegate some of their tasks to fog and cloud devices. Task offloading decisions are influenced by several factors, including device capabilities, network conditions, and application requirements. Most research work on task offloading problem has focused on only one objective, with few studies exploring the combination two or more objectives. Present work proposes a TOPSIS algorithm for ranking devices based on four objective function values - deadline met, execution cost, execution time and energy required, and selects the best suitable device for offloading a task. If the best device can not fulfil the requirements of a task, as an alternative, a fog device is selected. Load of the selected fog device is reduced by applying a task migration, so that the task requirements may be satisfied. This method aims to reduce the number of tasks that miss their deadlines and minimize the total execution cost. Additionally, the technique shortens total execution time, enhances energy consumption, optimizes the remaining energy of IoT devices and improves overall resource utilization. Extensive simulation tests were conducted by varying the size of the workload. The proposed method is compared with standard algorithms and the experimental results demonstrate that it increases reliability of meeting deadlines and fog device’s utilization, and reduces execution cost, execution time and energy consumption significantly.</p>