Improvement of real time and artificial intelligence based fault prediction system of electricity distribution network of Turkey with Industry 4.0 integration: Izmir-Cesme case study

World is going fast forward inside of a transformation in which there is continues flow of data between different systems, systems can alert each other, operational decisions and improvements can be executed autonomously by reason of means of developments in fields of artificial intelligence, hardware and software. Thanks to smart transformation, named as Industry 4.0, products, things, processes, machines, equipments and many more platforms are connected to each other in real time and again operations are executed in real time; daily life is affected in many aspects. Along with increase of Industry 4.0 integration level in Turkish public and private sectors, not only effectiveness of state governance and operation will increase; there will be an increase of quality in services of institutions give. There will be improvements of cost savings in State cost sections. In this study, together with integration of Industry 4.0 technologies gains in electricity distribution networks; hybrid model deep learning method for predicting hourly system status of distribution network and goal programming technique is suggested. In this study, status of distribution network and energy consumption in Çeşme and hourly data meteorological factors of the district are used. Prediction results of one hundred different deep learning models are obtained. The prediction results of deep learning models are used in the goal programming model. In terms of access to energy without interruption, it is aimed to increase the quality of service provided to individuals, public and private sector enterprises and to choose deep neural network model that minimizes the operating cost of the electricity distribution enterprise. An algorithm is developed that enables deep learning and goal programming models to work in real time, and suggests optimal prediction model. Study is featured as preliminary work in terms of its application in established and planned to be established electric distribution networks in Turkey by developing techniques in use with increase in number of factors that affect networks and increase of data volume. Keywords: Industry 4.0, Artificial Intelligence, Deep Learning, Goal Programming, Optimization, Operations Management, Electric Power Distribution Network, Fault Diagnosis, Production and Service Systems, Public Sector, Reliability