Generating Industrial Big Data as a Result of Implementing in Manufacturing Sectors and Automobiles

Big data can be generated in manufacturing sectors and automobiles by using Internet of things (IoT) technology where generation of myriad data is possible. Industrial IoT inspires the companies to change and adopt to new and emerging data-driven strategy. This paper explains about how IoT in manufacturing sectors and automobiles, will generate and store industrial Big data.

Introduction

In the fourth industrial revolution IoT and Big data plays a vital role as the manufacturing systems are transforming into digital ecosystems. IoT, the network of interconnected devices which exchanges data and thereby creating opportunities for increasing efficiency, reduce error and economic benefits. The exchange and storage of data in IoT will directly feed to big data which can be further used in a useful manner. Now a days many automobiles, machine components are equipped with IoT sensors to generate Big data. In modern and advanced industries, the data generated by IoT sensors is already being received in huge volume which is more than thousand Exabytes annually and its predicted to be even more in the upcoming years. These data-driven strategies will enable the companies to optimize their cost, errors and thereby increase the profit. The Big data generated will enable the company to work on predictive analysis and increase the competitive advantage in the market. This paper explains about how adopting IoT in manufacturing will generate Industrial Big data and how it can be further used in an useful manner with some suitable case studies and in addition to it cost optimized and non-disruptive IoT application for SMEs is explained by exploring the high volume of data which can be generated.

Literature review

D. Mourtzis (2016) has explained about the use of Industrial IoT to generate Industrial Big data and briefly discussed about the pros and cons. He finally concluded and analyzed about the amount and size of data that can be generated and analyzed in a case study of a shop floor of hundred machines. He had also done Mapping OPC-UA to the Open Systems Interconnection (OSI) model elaborated on it. J. Ben Naylor (2007) has used the Industrial Big data in a beneficiary way by using the lean principle and identifying the place and time where an error could occur by using predictive analysis.

Pramudianto F(2015) has done a work on using IoT in controlling Industrial Robot and also monitored energy consumption of individual robots and optimized it by using algorithms. STM32W node platform is used for monitoring robot motion. And they have concluded the application of big data which was generated by using IoT and further on how it can be used in a beneficiary manner, such as monitoring and optimizing the energy consumption.

Methodology

Generating Big data by using IoT in manufacturing sectors: A survey by Batty et. al, had predicted that industrial big data has reached the total volume of size more than 1000 Exabytes annually. Comparing the big data generated from IT firms its very less however it tends to increase in the upcoming years. For this reason the data generated by using IoT in Industries are called “Industrial big data” not “Big data”. The ultimate aim of adopting IoT in industries initiate smart factories, in which the individual machines are interconnected with each other and connected to a network in order to achieve it the resource should be connected to internet directly or through external adapters.

As a result of it, the machine tool system will be converted and transformed into a cyber-machine tool system enriched with knowledge which was acquired from the data collected and analyzed. And the resource also contains human operators connected through internet by using mobile devices and thereby converting operators into cyber operators. And finally, the IT and business tools will be connected to network. The data collected from low level enterprises are very important as these data can be analyzed to get some meaningful information which are be used by the higher-level enterprise. A main challenge towards this transformation is the design and development of standard and secure communication protocols capable of interfacing existing systems and collecting and exchanging manufacturing data. An IoT application, supported by a WSN and designed upon a standard industrial communication protocol is described below, presenting how Industrial Big data can be generated.

Monitoring system via WSN

A monitoring tool organized in a wireless sensor network (WSN) is presented. The monitoring tool consists of a data acquisition (DAQ) device which utilizes split-core current transformers (CT) as current sensors, a closed-loop hall effect current sensor, as well as a camera. These sensors are selected in order to create a non-intrusive and easy to install application for monitoring the status of machine-tools. The proposed tool is designed as an add-on for the commercial machine-tools, rather than communicating with the machine controller. This decision is mainly driven by the fact that the lifespan of the industrial equipment can reach the 50 years, hence old machinery often do not have the required capabilities for connectivity. Therefore, special effort is required to transform each legacy controller into an IoT device.

Experimentation/modeling

Implementation of IoTs in Two different studies cases have been discussed in this paper which uses IoT to generate Industrial Big data which can be further analyzed and used in a useful manner.

Case studies

1. In VIT machining process lab: By using IoT sensor like WSN, the energy consumption of individual machines in lab can be monitored and if any abnormal amount of energy consumption is taking place then can be monitored from the big data collected. It can also measure the optimized process parameter in which the energy consumption is notably less when compared with conventional process. For example, conventional turning process can be performed in a lathe which is connected to an IoT sensor and from which the big data have been recorded, so the energy consumption of lathe can be calculated with and without the usage of cutting fluid. The cutting fluid plays a vital role in heat dissipation in a machining process, so there will always be an energy loss in the form of heat when cutting fluid is not used. But when cutting fluid is used these losses of energy in the form of heat can be reduced. So, by implementing IoT in the machines the amount of energy that can be saved can be easily monitored and calculated. There are 4 Lathes and 3 drilling and 1 milling machine is the machining lab, considering 8 machines in total and the data that can be generated when is machine is continuously running for a day is 2GB. Considering a shop floor with100 machines then the data generated will ne 204 GB which will generate 6 TB of data per month and a huge 72 TB of data per year. This collected big data will have almost all the details about the machines such as on time, off time, energy consumption and tool change. Fig.2 Volume of data generated in a shop floor Fig.3 Volume of data generated in a shop floor.
2. IoT to monitor fuel consumption of an automobile component: Inadequate amount of fuel for automobile will be a serious issue in future, so IoTs to monitor the fuel consumption of an automobile system will be very vital in future. A case study on VIT cabs if these IoT are fixed in them is stated below. Considering the mileage of the cab to be around 20 kmpl and for a single trip it covers 2 km and the charge they will be charging is 15. So the Original will be Rs 200 including the driver charges. 14 students are enough to attain the breakeven point. But the specific fuel consumption will depend on the load carried and the number of students carried will determine the torque and torque will determine the fuel consumption these kind of complex calculations can be easily computed by adopting IoTs in automobiles.

Results and Discussions

The IoT paradigm transform the industries into “cyberproduction systems” capable of being flexible and adaptive and fully aware on the production conditions. However, new way of filtering and processing the data should be considering in order to reduce the produced and transmitted data. At present, there are no systems which help to identify the exact mileage of the four wheelers and view them in an graph format. It monitors approximate usage of fuel by the vehicle and Mileage is not efficiently and accurately calculated. Cars can only tracks the speed and kilometre through meters but does not keep a record of it. It does not offer daily monitoring of the mileage. So in both the case studies we have discussed about the importance of IoT and Big data in manufacturing and automobiles and the ways that it can be generated are also explained, further more studies can be conducted on the efficiency and accuracy of these data.

Conclusion

The proposed work presents how the adoption of IoT paradigm in manufacturing will generate Industrial Big Data. Industrial Big Data compared with the size of Big Data reported by Google, or Cisco, is of lower volume, however, it tends to be increased the next years. Industries are facing a new era of IoT. New monitoring services and the concept of IoT that tends to transform the machine tools into “cyber-machine tools” and the human operator into a “cyber - operator” will generated high volume and variety of data. In addition to that, new industrial communication protocols, such as OPC-UA will empower the interface with existing IT tools and will enable quick and accurate communication. The proposed work shows how the IoT paradigm in a simple case of a company of 100 machine tools considering different types of sensors can produce data and can lead to Industrial Big Data.