Algorithms for a Smart Construction Environment

Abstract

Every building project is one of a kind. Unique are its design, location, collaborating people and organizations, time-frame, cost and many other aspects. Setting up smart services and applications that would support the building process is on the wish list of every construction company. Such applications, however, have time-critical requirements that must be addressed in the application designing process. There are four broad types of converging technologies that are currently used to build smart applications. These include the Internet of Things (IoT), Artificial Intelligence (AI), Cloud Computing and Blockchain. The EU-Korean research and innovation project DECENTER intends to play a pivotal role in the integration of such technologies in a new Fog Computing Platform. The DECENTER Fog Computing Platform would help address the resource requirements of smart applications and provide high Quality of Service (QoS) to various AI-based applications. Due to the various uncertainties of the Fog Computing environment, the project aims at also providing Service Level Agreements for the offered services, which include assurances, ranking and verification of Edge-to-Cloud deployment options. Our present work focuses on the use of stochastic methods, particularly, the Markov Decision Process to deliver a QoS model of the smart application in relation to the available Cloud computing resources. This work presents the algorithm, which is used to achieve high QoS of smart applications and its implementation considerations. Our findings support the understanding that dynamic Edge-to-Cloud computing environments require a stochastic approach to establish assurances for high QoS operation of the applications. It is shown that the required assurances can be expressed as a probability value for confidence, which is calculated by the stochastic method.