Nowadays, more and more energy is consumed for economic growth and also for populations increasing. So, we have to think about how we can utilize energy to make industrial companies or factories more profitable. Industrial customer’s facilities and equipment use electricity for processing, producing, or assembling goods, including such diverse industries as manufacturing, mining, and construction. Overall, these industrial sector usages more than one-third of electrical energy from countries total usages of energy. Therefore, it is necessary to give technological support to propose some effective strategies to promote the industrial energy utilization rate. Industrial energy consumption analysis and predictions can help us to make a decision so as to reasonably control all kinds of equipment for reducing energy consumptions rate or utilize energy in a systematic way to reduce the energy cost for getting more profit from industrial factories.

Industrial factory owners are also starting to realize that analysing and predicting the energy data with manufacturing data is so much essential thing for their companies or factory’s profit. For improving the energy utilization of an industrial factory, this is the best time to work as a technological supportive hand with industrial factories managing board to improve their energy utilization better. By working together, we can make our industry more profitable by giving valuable results and suggestions.

In this study, we have two types of dataset and the dataset were collected from Daewoo Steel Factory in South Korea. The first one is industrial energy consumption data and another one is industrial manufacturing dataset or productions dataset. This study deals with analysing total usages of 24-hours energy consumption for every month and also total usages of 24-hours energy for the year 2017 to find out the maximum usages of energy time duration in each 24-hours. The major focus is to findout the relation curve between the total usages and demand of energy consumptions for the year 2017 by hourly. We make a relationship curve between total usages of energy, different types of the material weight of manufacturing products for every day in the year 2017 and also make a relation curve between per-day usages of energy, demands of energy and total materials weight for manufacturing products for the year 2017. Finally, we use the correlation method to find out the relationship of usages of energy with different types of materials weight for manufacturing products and also to find out the high significant factor for most usages of energy.

Se-Hark Park has been published a paper about decomposition of industrial energy consumptions with an alternative method [1]. Yemane Wolde-Rufael has published a paper about disaggregated industrial energy consumption and GDP: the case of Shanghai, 1952-1999 [2]. Ramazan Sari and et al, has been published a paper to find out the relationship between disaggregate energy consumption and industrial production in the United States: an ARDL approach [3]. Zheng_Xin and et al, has published a paper an improved grey multivariable model for predicting industrial energy consumptions in China [4]. Luis M.Candanedo and et al, has been published a paper data driven prediction models of energy use in a low-energy house [5]. Yue-Jun Zhang and et al, has been published a paper about the decomposition of energy- related carbon emission and its decupling with economic growth in China [6]. ZZhi-Fu Mi has been published a paper on potential impacts of industrial structure on energy consumption and CO₂ emission: a case study of Beijing [7]. Yansui Liu and et al, has been published a paper about assessing the impact of population, income and technology on energy consumption and industrial pollutant emission in China [8]. Elias Duran has been published a paper on Analysis and decomposition of energy consumption in the Chilean industry [9]. Qunwei Wang and et al, has been published a paper about Measurement and decomposition of energy –saving and emissions reduction performance in Chinese cities [10]. Elham Delzendeh and et al, has been published a paper on the impact of occupants’ behaviours on building energy analysis: a research review [11]. Fabrizio Ascione and et al, has been published a paper about a new methodology of the multi-objectives optimization of building energy performance [12]. Jianli Pan, Raj Jain and et al, has been published a paper about an internet of things framework for smart energy in building: designs, prototype, and experiments [13]. Mehmet Yalcinkaya and et al, has been published a paper about patterns and trends in building information modeling (BIM) research: a latent semantic analysis [14]. Seema Narayan and et al, has been published a paper on an investigation of renewable and non –renewable energy consumption and economic growth nexus using industrial and residential consumption [15]. Qinghua He has been published a paper on mapping the managerial areas of building information modeling (BIM) using scientometric analysis [16].

In this study, there are two types of dataset are available which gained from Daewoo Steel Factory, South Korea. Among these two data sets, one is for steel factory energy consumption data set and another one is steel factory manufacturing or productions data set. From energy data set, we get twenty-four hour usages of energy, demands of energy, lagging reactive power of the current, leading reactive power of the current, percentage of leading power and also the percentage of lagging powers are available for every day of the year 2017. From manufacturing or productions data set, we get the details about materials and productions like date of manufacturing product, product name, materials length, materials weight, material size, product item, product length, product weight, product size product standard and also lots of information are available. For checking per-hour usages of energy for every day we analysis, 2017 energy consumption data. We analysis energy consumption data of the year 2017 to find out every hour usage of energy and determine the time period of the high amount of energy usages for every day and also to determine the time period of less amount of usages energy. Figure 1 shows for 24-hours usages of energy consumptions for every day of the steel factory in the year 2017. In this figure the x-axis shows for the time in hours and the y-axis shows for the total usages of energy in kWh of every day in the year 2017. From the figure, we can see that most of the days from morning 8 am to night 10 pm usages of energy consumptions is very high and some of days it’s going on till 11 pm and we also see that after 11 pm to till 8 am energy consumption is low.

From figure we can also see that in the year 2017 some of the day’s energy consumptions are very low and those days are mostly weekly holidays and government holidays for the factory. Based on the figure we can easily say that Daewoo steel factory working hours are from 8 am to 10 pm and from 12 pm to 1 pm for lunch break and from 6 pm to 7 pm for dinner break.

<Figure 2> shows the different types of material weight of manufactured products for every working day. From the figure, we can see that there are three types of product are manufactured in the Daewoo steel factory and the products are “sheet”, “skelp”, and “cyong”. In this figure, the red color points show for per-day total material weight for sheet productions, the green color points show for per-day total material weight for skelp productions, and purple color points show for per-day total material weight for cyong productions.

Based on two dataset, we analysed 24-hour usages of energy consumption for every day in the year 2017. We make total energy consumption data in four groups as per Korean season’s winter, spring, summer, autumn and then make one analysing curve to find out in which season this factory usage maximum or high amount of energy. We analysis total usages of 24-hours energy consumption for every month and also total usages of 24-hours energy for the year 2017 to find out the maximum usages of energy time duration in every 24-hours. We identify the relation curve between the total usages and demand of energy consumptions for the year 2017 by hourly. We make a relationship curve between total usages of energy, different types of the material weight of manufacturing products for every day in the year 2017 and also make a relation curve between per-day usages of energy, demands of energy and total materials weight for manufacturing products for the year 2017. Finally, we use the correlation method to find out the relationship of usages of energy with different types of materials weight for manufacturing products and also to find out the high significant factor for most usages of energy.

3.1 Calculating Usages of Energy for Product Items

Here, PU – Product of Usage of energy, TU – Total Usages of energy, TPM – Total Product Material weight, TM – Total Material weight.

In this sector, we can see the analysis results of 24-hours usages of energy consumptions for every day of the year 2017, find out the seasonal usages of energy, find out 24-hours usages of energy consumptions for every month of the year 2017. We can see the relationship curve between total usages and demands of energy for 24-hours of the year 2017. We can see the relationship between total usages, demands, and materials weight for every day for the year 2017. We usages correlations method to show the relationship of usages of energy with different types of materials weight for manufacturing products and also to find out the high significant factor for most usages of energy.

In this study, we analyzed energy consumption and manufacturing data sets to define the usages and demands of energy corresponding to the weights of the materials of manufactured products. We found every month total usage of energy for every hour, see the seasonal usages of energy, the total usages and demands of energy for every hour in the steel factory in the year 2017. We see the relation curve between per-day usages of energy and different types of materials weight for different types of manufacturing product and see the relation curve between per-day usages of energy, demands of energy and total materials weight for manufacturing products of steel factory for the year 2017. We see the relationship of total usages of energy with different types of materials weight for manufacturing products in the steel factory.

This paper is focused on Industrial energy consumption analysis and prediction to improve the energy utilization rate to make profitable things for steel factory. From this study, we know about the working hours of the steel factory, seasonal usages and know about the correlations between total usages of energy and different types of manufacturing products. Factory working hour from 8 a.m. to 10 p.m. and this time factory maximum energy. Seasonal energy shows summer and winter time factory usages energy is very high then other two seasons. From correlations histogram, we can see that skelp productions are correlated with usages of energy better than other products. From <table 1> we see that using parentage of energy is for sheet producing 39.70%, for producing skelp 49.82% and for producing cyong 5.62%. This factory used maximum energy for producing skelp in the year 2017. On the other hand In Korea electricity cost is calculated in three types of daily rates and the three daily rates exist like off-peak, on-peak, and mid-peak. Customers are also offered different voltage levels, low voltage (220V and 380V) and three voltage levels ranging from 3.3kV to 345kV. We see a major price difference exists with peak rates, on-peak rates, and off-peak hours. Peak hour and mid-peak hour energy rate is higher than off-peak hour energy rates. So, for Utilize energy to make a profitable thing for factory, our suggestions are to run two shift in the factory for day and night time like in peak hours and off-peak hours than energy cost will reduce for this factory. As we find out skelp production is highly correlated with usages of energy, we can manufacture skelp in night shift.