Abstract
Internet of things (IoT) systems are taking an important role in daily life. Each year the number of connected devices increases considerably, and it is important to keep systems working appropriately. There are some options related to decision support systems to perform IoT systems tasks such as deployment, maintenance, and its operation on environments full of different connected devices and IoT systems interacting among them. For the decision-making process, the authors consider the complexity nature observed in IoT systems and their operational context and environments. In this sense, rather than using grain and fixed control rules/laws for the system design, the use of general principles, goals, and objectives are defined to guide the system adaptation. This has been referred to as guided self-organization (GSO) in the literature. The GSO design approach is based in evaluating the system entropy to reduce the emergence and enable self-organization. Also, in this chapter, a series of study cases from different IoT application domains are presented.
TopBackground
Decision-making occurs continuously in all scenarios of daily life; from the moment humans get up to start a new day until they go to rest; everything humans do in their daily lives is full of decision-making. Deciding what to eat, the route to follow to get to work, what tasks to carry out, the priority of each one, what clothes to wear, among others, are decisions that are made throughout the day.
Key Terms in this Chapter
Internet of Things (IoT): It is an infrastructure composed of one or multiple networks interconnected and composed of a set of elements with different properties that interact by collecting data and by means of actuators to achieve a set of previously specified goals.
Knowledge: Product obtained from information processing; it is used for future processing into the IoT system.
Guided Self-Organization (GSO): Enable Self-Organization taking into consideration the level of chaos of the system and its given goals.
Uncertainty: Level of ignorance that users have about the processes that a system performs.
Autonomic Computing: Includes self-sufficient computer systems, that is, capable of managing themselves using a set of objectives and the knowledge generated in real time of execution to know what to do in each situation.
Objectives: Set of guidelines assigned by the administrator that the system must follow to fulfill the tasks it must do.
Information: Product obtained from data processing, in which a meaning and purpose is given to each data.
Utility Theory: Allows to analyze the possibilities that a system has to carry out a process and to know the level of benefit or prejudice that said change will bring with it when taking one or another action.
Emergent Behaviors: Series of behaviors that arise constantly in a complex system, many of these changes occur unexpectedly.
Cyber-Physical Systems (CPS): Systems those integrate the physical part with the computational part and expand the capabilities of the physical environment in which they interact with the support of a network of sensors and actuators.
Self-Organization: Dynamic process belonging to complex systems in which the system changes behaviors to perform one or more tasks.
Entropy: It is a unit of measurement of the chaos and the content of the information available in a given system.