In recent years, various studies based on cyber physical systems (CPS) that integrate networking, computation, and physical processes have been actively carried out in different industries, national defense, and daily living. To physically reflect the theoretical aspect of CPS, consideration of various features is necessary to more easily integrate and effectively manage real-world components and the cyber world. This study suggests an adaptive resource management scheme (ARMS) to reduce the loss of sensing information and increase the level of accurate data obtained in the controller manager (CM) among the CPS components. A CPS-based system consists of a number of nodes (sensors and actuators) used to observe or monitor specific areas. ARMS utilizes data about the location and remaining battery capacity of each node to reduce the loss of information due to the irregular lifespans and unexpected breakdowns of resources in the CPS, and to obtain accurate data. Once a broken sensor in the physical world is sensed in the cyber world, the CM searches the locations of adjacent alternative nodes within a user-defined range based on the location of the broken node. In this process, an adjacent node search (ANS) algorithm is run to decide on a node (senor or actuator) to replace the broken node, taking into account the remaining battery capacity of candidate nodes. ARMS provides the adaptive resource management function of CPS by sending information on the identity (ID) and destination of the selected node to the controller to move the node to the destination and control the move.