The swift expansion of the information technology (IT) industry has led to a surge in compute-intensive and latency-sensitive applications. While cloud computing can satiate the demands of such applications, its centralized architecture may cause delays in the execution of tasks. To address such issues, edge computing brings computation closer to data sources. However, limited resources on Internet of things (IoT) devices make local execution quite challenging. Therefore, a pliable approach is to consider task offloading for moving heavy tasks to resource-extensive systems like edge/cloud. Osmotic computing, leveraging edge and cloud resources, aims to enhance IoT services. However, the dynamic nature of IoT, edge, and cloud introduces challenges for task offloading. This paper proposes an offloading algorithm using fuzzy logic to manage uncertainty. Furthermore, we introduce an osmotic decision manager (ODM) that employs fuzzy logic for optimized offloading decisions, considering IoT/edge for latency-sensitive tasks and cloud for latency-tolerant tasks. This algorithm aims to improve overall system performance by efficiently offloading tasks based on their specific requirements and constraints. The proposed algorithm undergoes simulation and assessment with diverse synthetic test cases to demonstrate its efficacy.