0
Research Papers

An Internet of Things-Based Monitoring System for Shop-Floor Control OPEN ACCESS

[+] Author and Article Information
Dimitris Mourtzis

Laboratory for Manufacturing
Systems and Automation,
University of Patras,
Patra 26500, Greece
e-mail: mourtzis@lms.mech.upatras.gr

Nikolaos Milas

Laboratory for Manufacturing
Systems and Automation,
University of Patras,
Patra 26500, Greece
e-mail: milas@lms.mech.upatras.gr

Aikaterini Vlachou

Laboratory for Manufacturing
Systems and Automation,
University of Patras,
Patra 26500, Greece
e-mail: vlachou@lms.mech.upatras.gr

1Corresponding author.

Contributed by the Computers and Information Division of ASME for publication in the JOURNAL OF COMPUTING AND INFORMATION SCIENCE IN ENGINEERING. Manuscript received October 3, 2017; final manuscript received February 8, 2018; published online March 15, 2018. Editor: Satyandra K. Gupta.

J. Comput. Inf. Sci. Eng 18(2), 021005 (Mar 15, 2018) (10 pages) Paper No: JCISE-17-1207; doi: 10.1115/1.4039429 History: Received October 03, 2017; Revised February 08, 2018

With the advent of the fourth industrial revolution (Industry 4.0), manufacturing systems are transformed into digital ecosystems. In this transformation, the internet of things (IoT) and other emerging technologies pose a major role. To shift manufacturing companies toward IoT, smart sensor systems are required to connect their resources into the digital world. To address this issue, the proposed work presents a monitoring system for shop-floor control following the IoT paradigm. The proposed monitoring system consists of a data acquisition device (DAQ) capable of capturing quickly and efficiently the data from the machine tools, and transmits these data to a cloud gateway via a wireless sensor topology. The monitored data are transferred to a cloud server for further processing and visualization. The data transmission is performed in two levels, i.e., locally in the shop-floor using a star wireless sensor network (WSN) topology with a microcomputer gateway and from the microcomputer to Cloud using Internet protocols. The developed system follows the loT paradigm in terms of connecting the physical with the cyber world and offering integration capabilities with existing industrial systems. In addition, the open platform communication—unified architecture (OPC-UA) standard is employed to support the connectivity of the proposed monitoring system with other IT tools in an enterprise. The proposed monitoring system is validated in a laboratory as well as in machining and mold-making small and medium-sized enterprises (SMEs).

FIGURES IN THIS ARTICLE
<>

Manufacturing enters a new era, where higher levels of flexibility are required to address the challenges for shorter product lifecycles, increasing number of new products and variants, uncertainties, and fluctuations in the market demands [1] especially for addressing the needs for mass customization and personalization [2] often in decentralized manufacturing environments [3].

The pinnacle of computing technology in the 21st century with powerful and cost-efficient processors, storage devices that can contain many terabytes of information, along with autonomous embedded systems that are being wirelessly networked with each other, enable the convergence of the physical and the cyber worlds. The procedure of digitalization of contemporary factories is not straight forward and depends highly on the nature of each manufacturing system. Nevertheless, the companies that are forward-thinking and confident into adopting innovative operational models will be able to endure the emerging market demands.

The advent of modern technologies such as cyber-physical systems (CPS), internet of things (IoT), and big data analytics open new horizons toward the industrial digitalization by enabling automated procedures and communication by means that were not attainable in the past. The usage of processors with high processing capabilities to embed intelligence into manufacturing resources transforms passive systems into active entities with decision support capabilities. This transformation is described in the paradigm of IoT [4]. However, the application of IoT, especially in industry, results into the creation of vast amounts of heterogeneous information that needs special manipulation and analysis to perform meaningful reasoning and extract the actual value. The extraction of the knowledge from the data collected in all levels of manufacturing systems can create autonomous smart manufacturing systems [5].

To leverage the modern technologies toward the digitalization of contemporary manufacturing systems, this paper presents a monitoring system for machine-tools following the IoT and CPS paradigms. The system includes the collection of information from data acquisition devices (DAQs) on machine-tools. The DAQs employ sensors and are organized into a star wireless sensor topology with a microcomputer being the coordinator. Data processing is performed inside the DAQs and the microcomputer to extract meaningful information and transmit it to a Cloud server for aggregation and visualization. The outputs of this monitoring system are referred to the actual status of the resources, the electrical energy consumption, and information related to the machining tasks that are performed. As a result, the proposed monitoring system can support manufacturing companies to perform adaptive and efficient decision making based on the actual shop-floor condition.

The structure of the paper is organized as follows: Section 2 deals with the literature review on the technologies and the previous approaches related to the proposed monitoring system. Section 3 describes the proposed system. Section 4 presents the hardware developments, while Sec. 5 presents the microcomputer software developments. In Sec. 6, the case study is presented and in Sec. 7 the results. Finally, Sec. 8 concludes the paper.

In this section of the paper, a literature review on the key enabling technologies for the proposed monitoring system is performed. The literature review covers the subjects of the cyber-physical systems, the internet of things, the shop-floor monitoring, and the cloud manufacturing.

Cyber-Physical Systems.

Cyber-physical systems have been defined as “the systems in which natural and human made systems (physical space) are tightly integrated with computation, communication and control systems (cyber space)” [6]. They link the physical world seamlessly with the virtual world of information technology and software [7]. They employ various types of available data, digital communication facilities, and services [8]. The evolution of manufacturing science and technology leads to the adoption of CPS in industry, also known as cyber-physical production systems [9]. Adaptive and collaborative process planning using function blocks and Web technologies has been proposed in Ref. [10]. The modeling of the CPS for manufacturing systems can follow well-known system description frameworks or extending existing such as the EAST-ADL modeling language [11]. The concept of cyber-physical production systems applies also in the resources as a low-level CPS modeling [12]. Despite the connection with the tangible resources, the CPS can be extended with the social media usage in industry toward Social Manufacturing [13]. A detailed review on the CPS in manufacturing captured from the quality of service perspective has been performed in Ref. [14].

Internet of Things.

The cyber instances in a CPS can be regarded as virtual objects with storage and processing power, that communicate with other cyber entities and humans, and can control the physical device on which they are attached. These virtual objects are following the paradigm of IoT. The feasibility for the adoption of IoT devices is also based on the reduction of the battery prices and the increase of their energy density [15]. The use of sensors and the computational capabilities of modern microcontrollers are complemented by the communication capabilities that embedded systems provide. Moreover, the information networks that are based on the IoT can create new business models, improve business processes, and reduce costs and risks [15].

The IoT is a multidisciplinary field [16]. The identified challenges include the standardization, issues related to network bottlenecks, the security of the communication, and the intellectual property protection [16]. The huge amount of data that are generated by IoT require huge processing capabilities. Therefore, the context-aware computing methodologies will facilitate the decision making on which data will be processed and in which layer [17]. An important part of the IoT networks are the gateways, as they are responsible for the management of the local network of devices [18]. Moreover, middleware and cloud platforms have been evaluated to integrate the IoT devices and host the generated data [19]. Nevertheless, the existence of numerous Web-services that can be used for IoT applications requires a methodology for predicting the quality of these services and support the engineers selecting the appropriate [20]. Successful applications of IIoT that enhance the value for the customer in the subjects of monitoring and control, business analytics, information sharing, and collaboration have already been evaluated in real-world case studies [21].

In manufacturing, the wireless sensor network (WSN) topologies are the most eligible candidates to facilitate IoT communications, as they offer flexibility and scalability [22]. A WSN consists of many wireless capable sensor devices working collaboratively to achieve a common objective through standards based on the requirements of each specific application [23]. For the local communications of WSNs, the ZigBee standard has gained much attention lately [24].

Shop-Floor Monitoring.

The monitoring and data collection during the manufacturing operations is the foundation for factory automation and decision making. The CPS paradigm suggests the use of monitoring devices under the IoT philosophy that go beyond the traditional approaches for on-site data collection, processing, and visualization. The necessity for the use of real-time monitoring in manufacturing has been stressed in the work of Teti et al. [25]. In the context of monitoring machine-tools, there is a variety of sensors that can be employed (acoustic, vibration, force, current, etc.) [25]. A less frequent sensing method employs visual sensors for monitoring the performance of a production system [26]. Electrical current sensors are promising candidates for measuring energy-related operating characteristics, since they are cost efficient and nonintrusive in nature [27]. In previous literature, monitoring systems have been proposed for the purposes of preventive maintenance [28], remaining useful life estimation [29], and cutting tool reconditioning [30] among others. The application of monitoring devices in the shop-floor to track the availability of machine tools, results in an adaptive holistic scheduling and has been proposed in Ref. [31]. The awareness on the shop-floor situation through monitoring systems has facilitated also adaptive process planning in Ref. [32].

Cloud Manufacturing.

The philosophy of cloud manufacturing can act as enabler for data exchange between IT tools and ubiquitous access to information by multiple users and software [3335]. Τhe adoption of cloud technology in manufacturing is strongly related to the quality of the provided services [36]. The capabilities for scalability according to the fluctuations in the demands for services and the flexible licensing models are major benefits of cloud manufacturing [37]. Moreover, cloud manufacturing implies an integrated cyber-physical system that can provide on-demand manufacturing services, digitally and physically, at the best utilization of manufacturing resources [38]. In addition, the semantic representation of information can facilitate collaboration through the service-oriented architecture, especially in distributed manufacturing environments [39].

Considering the previous literature and the industrial practice in the aforementioned fields, several approaches are reported related to shop-floor monitoring. However, novel, robust, reconfigurable, reliable, intelligent, and inexpensive monitoring systems is a field of further investigation to meet the demands of advanced manufacturing technology [25]. Toward that end, the proposed work presents an inexpensive, reliable as well as easily reconfigurable cloud-based monitoring system for raising the awareness on the shop-floor condition and the utilization of resources, supporting adaptive and efficient decision making. The proposed monitoring system consists of a developed data acquisition device together with a wireless sensor network for data capturing instead of interfacing the controllers of the machine-tools, to result into a plug-and-play approach that is suitable even for legacy resources to advance into the IoT era. This facilitates the transformation of contemporary shop-floors into CPS and their benefit from digitization technologies. The cloud is selected in this work to host the central server, as it is an enabling technology for the “always online” operation of the software and the potential for flexible licensing models.

This study aims to present a monitoring system for smart shop-floors from the IoT perspective. The main objective of the proposed system is to improve the awareness on the actual status of the manufacturing resources and especially machine-tools. It is rather common in industry to create production schedules, while taking into consideration resources that might not be available when required due to various reasons. Hence, the issue of the awareness is of crucial importance to avoid bottlenecks and increase productivity.

The work presented in this paper is part of an integrated system including the Hardware and the Software modules to perform IoT-based monitoring. This paper focuses on the hardware developments and the software that is related to the connection of the manufacturing resources to the cloud. The presentation of the software of the cloud server can be found in Ref. [31].

Overall Architecture of the Monitoring System.

This monitoring system consists of DAQs, a wireless sensor network, and a cloud server. There is one DAQ for each machine tool. On each DAQ, the sensors that measure the operating characteristics of the corresponding machine-tool are connected. The DAQs of a shop-floor communicate with a microcomputer via wireless communications in a wireless sensor network. Therefore, the sensors are connected on the DAQ through wires and the DAQ is responsible to preprocess the sensor outputs and transmit the measurements in the wireless sensor network. Considering that electrical power is always available for the operation of the machine tool, the DAQ meets its energy requirements through the power supply of the machine-tool. Hence, no additional wiring is needed. The outputs of the system are intended to be transferred to a cloud server for visualization (Fig. 1).

As mentioned before, the proposed system 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 30 years [40]; hence, old machinery often do not have the required capabilities for connectivity. Therefore, special effort should be made to transform each legacy controller into an IoT device.

Another pillar for the proposed architecture is the distributed data processing. Each sensor may produce a large amount of data which can result into some gigabytes per day [41]. It is evident that the unprocessed data streams do not represent meaningful information. Hence, to extract meaningful information from the raw data that will result into knowledge on the monitored entity, the appropriate processing needs to take place. In this work, the processing and the reduction of the data is performed on the source of its generation, i.e., the sensor. This is achieved through the microcontrollers of the data acquisition devices that are installed in the machine tools.

In addition, the connection of each individual IoT device directly to the internet may result into communications bottleneck, due to lack of sufficient bandwidth. For this reason, each IoT device (i.e., the DAQ) connects to a gateway via a local wireless sensor network. The gateway is the coordinator of the network and its objective is to transmit the meaningful information of each machining task to a central repository.

Each layer of the proposed architecture follows the idea behind IoT and provides integration aspects for the realization of interoperable systems. The DAQ architecture allows the usage of more sensors and communication protocols; the gateway provides interfaces with industrial networks, in this study open platforms communication—unified architecture (OPC-UA) is considered and an information model is presented; the higher level of the central server supports integration with industrial software such as enterprise resource planning and manufacturing execution system.

Sensors and Data Capturing.

The sensors for the status identification of the machine-tools measure the overall electrical power consumption of the system and the individual current consumption of each one of the main motor drives. Specifically, the individual motor drive measurements include current sensors on the spindle and on each one of the moving axes. The sensors for the axes drives are split-core current transformers. Especially, for the case of the spindle, a close-loop hall-effect current sensor is selected to capture the overall harmonic content of the spindle current. The overall electric power consumption of the machine tool is measured through a current sensor installed in one of the three mains lines. The measurement of one line is preferred instead of measuring all three lines, since the machine-tools act as balanced loads. Therefore, Eq. (1) applies without errors. The outputs of the current sensors are driven to the microcontroller after the required signal manipulation. Moreover, the voltage of the mains is measured through an insulation transformer for safety reasons. To identify correctly the status of the machine, a proximity switch is installed in the door of the machine-tool, which is open in the case of the setup mode Display Formula

(1)P=3·Vph·Iph·cosφ

where P is the active power (W), Vph is the phase voltage (V), Iph is the phase current (A), and cosφ is the power factor of the three-phase load.

The outputs of the current transformers are sampled by a frequency of 1 kHz which corresponds to 20 samples per period (in the case of 50 Hz). For these current measurements, only the root mean square (RMS) values are calculated through the following equation: Display Formula

(2)IRMS=1n1nin2

where in is the sensor measurement (Α) and n is the total number of measurements.

For the close-loop hall-effect current sensor, the bandwidth is 200 kHz; therefore, the maximum sampling rate was set to 1 MHz. Except for the RMS value, which is also calculated following Eq. (2), the fast Fourier transform is also calculated for the hall-effect sensor. Hence, the signal is transformed from the time domain to the frequency domain aiming to extract the harmonic content of the current. When digital sampling is applied to analog signals, the discrete Fourier transform (DFT) is used to extract the spectrum of the signal (Eq. (3)) [42]. The fast Fourier transform is a very efficient algorithm for computing DFT coefficients Display Formula

(3)X(k)=n=0N1x(n)ej2πknN,k=0,1,,N1

where X(k) — the DFT coefficients, x(n) is the periodic digital signal after the sampling of its analog counterpart and N is the acquired data samples.

This functionality was considered while designing this monitoring system, driven by the fact that, in every electric motor, the mechanic torque is closely related to the motor current [43]. Thus, observing the harmonic content of the motor current, useful insights that are related to maintenance aspects can be deduced. The full implementation of this functionality paired with the necessary pattern recognition methodology to identify deviations in the spindle current spectrum that are related to mechanical faults will be demonstrated after the long-term installation of the system in machining industry. This will provide the amount of data required to conclude into accurate results.

Wireless Sensor Network Design.

The DAQs of a shop-floor are organized in a WSN following the star topology. The selection of the WSN was driven by the requirements for flexibility and reduced infrastructure. The data transmission is coordinated by a central gateway which is responsible to collect the data from the DAQs and organize them into packets before transmitting them to a cloud server for further processing and visualization.

The WSN is facilitated with the use of DIGI XBee ZigBee RF module. ZigBee is a specification of the IEEE 802.15.4 standard. The selection of ZigBee over other wireless standards is performed due to its support to various network topologies and encryption algorithms, and its robust operation with functionalities such as collision avoidance, retries, and acknowledgements performed in the hardware. Moreover, ZigBee modules can communicate in ranges more than 100 m [44].

The data within the WSN are transmitted as ZigBee frames that have unique recipients. To automate the addition and the removal of the nodes in the WSN the following procedure is developed (Fig. 2). In the first step, each DAQ node transmits a beacon message once every 5 s. If a DAQ is in transmission range of a coordinator, the coordinator receives the beacon message and verifies the DAQ address with a list of registered DAQs. If the DAQ address is registered in the coordinator, the coordinator transmits an “initiate communication” frame. Then, the DAQ abandons the beacon mode and waits for the coordinator to request a measurements packet. Subsequently, the coordinator requests the measurements of each DAQ once every 0.25 s and the operation of the network continues following this manner.

To avoid network malfunctions due to problematic devices or absent nodes, supervisory mechanisms are implemented into both DAQ and coordinator devices (Fig. 3). The coordinator sets a specific flag when a request for packet is sent to each DAQ. If the DAQ fails to reply before the beginning of the next cycle of requests by the coordinator, the latter adds the value “1” to a scorecard of the corresponding DAQ. In the opposite occasion of the successful reply by the DAQ, the coordinator subtracts the value “1” of the sum in the scorecard. If the score of each DAQ reaches the value of 20, the coordinator perceives this node as offline and stops requesting the corresponding measurements. On the other side, the DAQ which is not in the status of beacon and communicates with a microcontroller monitors the presence of the coordinator following a similar algorithm. The DAQ has a scorecard for the coordinator and adds into the sum the value “1” if the coordinator does not send a request for measurement in the expected timeframe of 0.25 s. For a successful receive of a request for measurements the DAQ subtracts the value “1” of the scorecard. After reaching the score of twenty, the DAQ considers the coordinator absent and re-enters the beacon mode.

Main Outputs of the Internet of Things-Based Monitoring System.

The proposed monitoring system provides information about the operation of the machine related to the machining tasks. This is achieved by synchronizing the physical with the digital world through the IoT-based DAQs. The main outcome of the system is that it increases the awareness on the resources by determining their actual status. The sensors identify whether the machine is “Processing,” “NonProcessing,” or in “Setup.” This information is useful in production scheduling activities where the knowledge on the actual availability of the resources is crucial to the design of feasible production schedules. Furthermore, the detailed information on the time periods that are required to perform specific tasks, along with the required setup times can make the future production schedules more accurate [31].

Moreover, the use of current and voltage sensors can provide accurate information on the electrical power consumption of the machine-tool and its independent drives. This knowledge can contribute toward the estimation of the electrical cost per product and the reduction of the environmental footprint of the production systems. As already discussed, the output of the hall sensor can give insights on failures prior to their occurrence. Furthermore, the calculation of the actual machining time of the machine-tool subsystems enables a more efficient preventive maintenance planning, instead of making maintenance tasks into fixed time intervals without considering the usage of the machinery [45].

Based on these outputs of the monitoring system, a relevant set of performance indicators (PIs) is selected. These performance indicators can be found in Table 1. Except for these PIs that can be directly obtained, this IoT based monitoring system can give low level information as an input to PIs that are referring in the higher levels of the production hierarchy. Such PIs can be aggregations of the PIs presented in Table 1 and refer to the production line, such as the PI “production line availability.” Moreover, composite PIs can be defined that require the measurement of more than one simple PIs and metrics.

The final step in the data transmission is the cloud server. The communication with the cloud server is performed for two different purposes, the first is the real-time streaming of the measurements and the second is the storage of the task related information for future reference. Both data streaming and the task reports are useful information for administrative purposes in the higher levels of the enterprise. In the cloud level, integration with other industrial software modules, such as enterprise resource planning and manufacturing execution system, can be performed. This facilitates the interoperability among various systems and the multidirectional information flow. In this context, Ref. [46] employ case based reasoning to estimate the electrical energy consumption required for the manufacturing of new parts, based on historical data.

The design and the development of the proposed data acquisition device is performed to achieve an inexpensive, reliable as well as a reconfigurable solution for the industrial companies. In addition to the data acquisition device, a main aspect of the proposed monitoring system is the connectivity. Different communication protocols are applied to the proposed monitoring system to support its connectivity and ensure the quick and accurate data transmission.

Specifications.

To achieve the objectives presented in Secs. 3.13.4, the DAQ has been developed following the industrial requirements for compact size and robust design, which have been enhanced through real-life demonstrations.

The selected split-core current transformers, which are used as current sensors, have a transformer ratio of 1:2000 and are rated to 100 A RMS nominal primary current. The closed-loop hall-effect sensor for the measurement of the spindle current is selected in comparison to an open-loop due to its immunity to temperature changes and the nonsaturation of its core; therefore, excellent linearity is achieved.

To comply with the appropriate specifications for supporting the selected sensors, communication capabilities, and computational requirements, the STM32F429 microcontroller from ST Microelectronics was selected [47]. The microcontroller has the ARM Cortex-M4 microprocessor at its core and a special processing unit for floating point arithmetic. The operating frequency of 180 MHz along with the special processing units give the capability for real-time signal processing on the DAQ, which is essential considering the sampling frequencies for the machine motor electrical operating characteristics.

Connectivity.

The connectivity requirements for the DAQ are separated in two layers. The first layer considers the communication of the DAQ with external systems. Hence, for the communication with the microcomputer coordinator of the WSN a XBEE ZigBee module is installed. Moreover, to support the on-site and high-speed data transmission for high resolution on the measurements, a USB interface is considered. In addition, to communicate with wireless sensors in small ranges or with mobile devices, a Bluetooth interface is also implemented. The second layer includes the communication of the microcontroller with the other hardware peripherals inside the DAQ. This includes the communication with the analog-to-digital converter via 1.4 MHz serial peripheral interface, the communication with the XBEE via universal asynchronous receiver/transmitter (UART) in 56,000 bps. UART in the same baud-rate is implemented also for the Bluetooth and the universal serial bus USB communication. Taking advantage of the openness of the designed DAQ, other types of sensors such as accelerometer and microphone are supported via UART, serial peripheral interface, or inter-integrated circuit (I2C).

The IoT philosophy considers interoperable systems with capabilities to support updates in an automatic manner. In this context, the digital input/output pins of the XBEE module are used to perform over-the-air updates on the microcontroller of the DAQ. This maintains the reconfiguration of the DAQ and the addition of features that will emerge as new requirements from industry as manufacturing evolves. In this local network, the data communications are secured using the advanced encryption standard 128-bit encryption algorithm. A block diagram representing the architecture and the components of the proposed DAQ is presented in Fig. 4.

The WSN is coordinated by a microcomputer which is responsible to collect all the data from the nodes of the network. These data are locally stored in the gateway and reports are created at the end of each task. These reports are then transmitted to the central server for administrative purposes.

Platform Selection and Software Development.

The selected microcomputer for this purpose is a Raspberry Pi 2 with a Linux operating system (Raspbian). The communication with the DAQs of the shop-floor is performed through a ZigBee XBEE module on an USB-to-serial converter. The measurements for each task are stored into a structured query language (SQL) database. The factory is modeled in a four-layer architecture that consists of the factory, the job-shops, the workcenters, and the resources (in our case the resources are the machine-tools) [1]. Each machining task is necessarily assigned to a machine-tool and each machine belongs necessarily to a workcenter. On the other hand, the measurements that refer to a machine-tool may belong to a task or correspond to the nonprocessing status. As soon as the DAQs are connected with the microcomputer, the data transmission is initiated. The measurements that are transmitted every 0.25 s from the DAQs are captured by the microcomputer and are stored in the database. The microcomputer adds a timestamp on each measurement, along with the machine status flag. The software of the gateway is composed using the Python programming language.

The status is determined by comparing the measurements with calibration values that correspond to each individual machine. Therefore, the calibration levels must represent the threshold of the idle current for each one of the monitored electric motor drives. If the measurement exceeds this level the status is set to Processing, else the status is set to nonprocessing. Moreover, if the proximity sensor that is attached on the door of the machine is triggered, the status is set to setup. At the end of the machining task, the microcomputer automatically performs a query to the database to retrieve the measurements related to this machining task. Subsequently, the necessary calculations are performed to create a task report that will be sent to the main server. The task report includes the timestamp of the start and the end of the task, the total energy consumption, the total setup time, the total processing time, and the total nonprocessing time. The task report is transmitted via hyper text transfer protocol requests to the main server which is deployed on cloud. The communication with the cloud server is secured using the secure sockets layer/transport layer security standard.

The selection of a cloud server provides ubiquitous access to information and flexible licensing schemes for the delivery of this monitoring system as a service [32]. The cloud server is hosted using an OpenSUSE Linux operating system. The Web application is developed using the Ruby-on-Rails framework. Detailed presentation of the software can be found in Ref. [31]. Furthermore, the microcomputer can stream the measurements directly to the cloud server for real-time visualization of the monitoring results. This functionality, is of great importance in the case of distributed manufacturing environments, where the production control may be on a different location than the actual production.

Open Platforms Communication—Unified Architecture Compatibility.

This IoT-based monitoring system is designed to support integration with existing industrial equipment. The OPC-UA is a standard that allows servers to provide real-time process data, environment metadata, and even nonprocess data to clients, in a unique and platform-independent way. The OPC-UA can provide communications in all levels of manufacturing enterprises, from the resource to the factory level, via a service oriented architecture.

Open platforms communication—unified architecture provides all data in its unified address space. It can represent anything starting from a simple variable to a complete machine. The OPC-UA provides an extensible data model, which provides the data schema. Hence, even systems that are not familiar with the data model can retrieve information from other systems. The client can browse through the server to gather the required information and understand its content through metadata and semantic representation. In the proposed system the microcomputer acts as an OPC-UA server with binary encoding in the transmitted data. The information model follows the specification [48] and is designed to correspond to the data stored into the SQL database.

In Fig. 5(a) reduced version of the model is presented for simplicity, which does not include the job-shop and workcenter objects along with the low-level variables and variable types. The main objects of the information model are the shop-floor, machine, task, spindle, axis, and measurement. The relationships among them are defined through the references “HasComponent.”

The proposed monitoring system is validated in a laboratory, a Spanish machining small and medium-sized enterprise (SME), an English machining SME as well as in a Greek mold-making industrial SME. The proposed monitoring system was installed in all the aforementioned cases and was improved following the feedback from each case. All functionalities were tested in real manufacturing environments except for the Hall effect sensor of the spindle, which will be evaluated in future work along with a pattern recognition methodology. Nevertheless, the relevant signal processing firmware for the microcontroller was evaluated using test signals.

The first evaluation was performed in a Spanish and an English SME. In this evaluation, a first version of the DAQ was designed and applied. Following the specifications of the two different machine tools from the two SMEs, the sensor board was designed and applied in the two cases measuring the current of the main motors and the overall power consumption. In this application, also the wireless sensor network was first validated transmitting data from the data acquisition devices to a central micro-computer through XBee ZigBee RF module, which is a specification of the IEEE 802.15.4 standard. The second evaluation was performed in a Greek mold-making SME. In this case, the data acquisition device was applied in two different machine tools and the monitoring data were transmitted through the wireless sensor network (ZigBee) to a microcomputer gateway and then through hypertext transfer protocol request to a cloud server that the company utilized.

The evaluation of the last version of the DAQ was performed in a laboratory environment, on an XYZ SMX SLV 3-axis machine-tool. The developed DAQ was installed in the electrical cabinet of the machine and measurements were captured for machining tasks of short and long duration respectively. In this case also the functionalities of the OPC-UA standard were tested and validated. The measurements were stored in the database in a rate of two measurements per second.

The results from the short-duration machining task are presented in Fig. 6. The processing of the monitored data is performed in the gateway layer by dedicated software. The software that manipulates the data transmitted by the DAQs and is responsible for the task reports is composed using the Python programming language. This software performs SQL queries to the database and generates the results that are presented in this section. Nevertheless, there is the option to stream unprocessed data values to the cloud server for ad-hoc operations and results. The duration of this task was approximately 3.5 min and included a brief milling operation.

In Fig. 6, the discrete events that are associated with the variations in the level of the power consumption are identified and marked. The first event that corresponds to a very high peak in the apparent power is the acceleration of the spindle. This high peak is justified by the fact that an induction motor consumes approximately 7 times higher than its nominal values of current when accelerated from zero revolutions per minute. Following the spindle peak, the power consumption rises a little when the positioning of the Z-axis is performed. During the material removal process, a rise in the overall power consumption can be observed. Finally, a peak, but significantly lower than the one of the spindle, can be observed when positioning of the axes is performed in rapid feed. From the graph, it can be concluded that even though in the selected machine tool the peripherals account for a little portion of the overall power consumption, the portion that is related to the actual machining process is very small. The largest portion of the energy consumption in this machine tool is related to the spindle movement. Subsequently, the DAQ was collecting measurements from the machine-tool for a whole milling operation to correlate the measurements with the actual status of the resource. The operation lasted approximately 130 min and the measurements that correspond to the apparent power consumption and the current of the spindle motor are presented in Fig. 7. The discrete events of the operation are identified and marked in Fig. 7.

The task report that is generated from the IoT-based monitoring system and refers to the long-term milling task that is depicted in Fig. 7 and the corresponding data are presented in Table 2.

Since the advent of thousands of IoT devices may result into storage bottlenecks [41], the data that are generated and stored in each level of the proposed monitoring system, referring to the operation of one machine, are depicted in Table 3.

The electrical current consumption of the developed DAQ was measured of an RMS value of 90 mA which corresponds to 2.16 Watts of power consumption using a power supply providing 24 V DC. The low power consumption is resulted due to the low power consumption of the ST Microcontroller and the selected peripheral devices. Moreover, the power supply electronic circuits that have been developed are switching DC-to-DC converters, instead of linear voltage regulators that are characterized by their low energy efficiency.

The proposed work presents a machine-tool monitoring system based on the IoT paradigm. The main objective of the system is to raise the awareness on the actual status of the manufacturing resources and support integration with existing industrial systems via the OPC-UA communication standard and Internet Web-services, aiming to perform adaptive and efficient decision-making. The system consists of three layers i.e., the DAQs with the sensors in the machine-tools, the microcomputer gateway with the local database, and the central cloud server. Moreover, a wireless sensor network is implemented to transfer from the sensor nodes to the microcomputer coordinator. The developed framework provides automatically reports on the tasks performed in a shop-floor, which is a procedure that up until now included a lot of manual work. This provides new capabilities for manufacturing companies to advance into the digital era and harvest the benefits that arise. In addition,, the proposed monitoring tool is designed and developed to be inexpensive, reconfigurable as well as reliable. Moreover, the developed system enables the control of distributed manufacturing environments through the Internet and cloud technologies.

The benefits of the digitalization of manufacturing lie in the increasing of the resources utilization, reducing the total machine downtime, increasing the productivity by automating the knowledge work, reducing inventory and quality related costs, increasing forecasting accuracy, reducing the time to market and maintenance costs [49]. Toward these ends, the proposed framework can contribute to the increase of the resources utilization by identifying their actual status, reduce the machine downtime through condition-based preventive maintenance and predictive maintenance by processing of the data gathered from the sensors [45], and knowledge reuse through machine-to-machine communications and information retrieval from the cloud server [46].

Other benefits of the system are the easy installation in the electrical cabinet of the machine-tools and its nonintrusive nature, as it does not require modifications in the electrical wiring of the machine. Moreover, the selected sensors and electronics are cost efficient, a fact that enables the purchase of the DAQ and the gateway with low investment. The low cost, along with flexible licensing models that are facilitated by the cloud technologies make the system ideal for SME manufacturers, providing them with inexpensive and reliable solutions.

In the future work, the authors intend to present in detail the cloud server with its corresponding software that visualizes the data captured by the sensors, and communicates with the scheduling module of a planning system and the operators of the machine-tools. All this information from the heterogeneous sources will be fused through an information fusion technique to enhance the captured results. Finally, specific focus will be given to the correlation of the harmonic content of the spindle current with potential failures of machine tools.

  • Partially supported by the EU funded research project “Collaborative and Adaptive Process Planning for Sustainable Manufacturing Environments—CAPP4SMEs” (No. 314024). The EU funded research project “Advancing Legacy Machine Tools into the Digital Manufacturing Century—LegInt” from the “CPS Engineering Labs—expediting and accelerating the realization of cyber-physical systems” (No. 644400).

Chryssolouris, G. , 2006, Manufacturing Systems: Theory and Practice, 2nd ed., Springer-Verlag, New York.
Mourtzis, D. , and Doukas, M. , 2014, “ The Evolution of Manufacturing Systems: From Craftsmanship to the Era of Customization,” Handbook of Research on Design and Management of Lean Production Systems, V. Modrák and P. Semančo , eds., IGI Global, Hershey, PA, pp. 1–29. [CrossRef]
Ilsen, R. , Meissner, H. , and Aurich, J. C. , 2017, “ Optimizing Energy Consumption in a Decentralized Manufacturing System,” ASME. J. Comput. Inf. Sci. Eng., 17(2), p. 021006. [CrossRef]
Löffler, M. , and Tschiesner, A. , 2013, “ Internet of Things and the Future of Manufacturing,” McKinsey & Company, New York, accessed Sept. 28, 2017, http://www.mckinsey.com/business-functions/business-technology/our-insights/the-internet-of-things-and-the-future-of-manufacturing
Feng, S. C. , Bernstein, W. Z. , Hedberg , T., Jr. , and Feeney, A. B. , 2017, “ Toward Knowledge Management for Smart Manufacturing,” ASME. J. Comput. Inf. Sci. Eng., 17(3), p. 031016. [CrossRef]
Bagheri, B. , Yang, S. , Kao, H. A. , and Lee, J. , 2015, “ Cyber-Physical Systems Architecture for Self-Aware Machines in Industry 4.0 Environment,” IFAC-PapersOnLine, 48(3), pp. 1622–1627. [CrossRef]
Sztipanovits, J. , and Ying, S. , 2013,“ Strategic R&D Opportunities for 21st Century, Cyber-Physical Systems—Connecting Computer and Information Systems With the Physical World,” National Institute of Standards and Technology (NIST), Gaithersburg, MD, accessed Sept. 28, 2017, http://www.nist.gov/el/upload/12-Cyber-Physical-Systems020113_final.pdf
Mikusz, M. , 2014, “ Towards an Understanding of Cyber-Physical Systems as Industrial Software-Product-Service Systems,” Proc. CIRP, 16(1), pp. 385–389. [CrossRef]
Monostori, L. , Kádár, B. , Bauernhansl, T. , Kondoh, S. , Kumara, S. , Reinhart, G. , Sauer, O. , Schuh, G. , Sihn, W. , and Ueda, K. , 2016, “ Cyber-Physical Systems in Manufacturing,” CIRP Ann.-Manuf. Technol., 65(2), pp. 621–641. [CrossRef]
Wang, L. , Shen, W. , and Lang, S. , 2004, “ Wise-ShopFloor: A Web-Based and Sensor-Driven e-Shop Floor,” ASME J. Comput. Inf. Sci. Eng., 4(1), pp. 56–60. [CrossRef]
Chen, D. , Maffei, A. , Ferreirar, J. , Akillioglu, H. , Khabazzi, M. R. , and Zhang, X. , 2015, “ A Virtual Environment for the Management and Development of Cyber-Physical Manufacturing Systems,” IFAC-PapersOnLine, 48(7), pp. 29–36. [CrossRef]
Chen, J. , Yang, J. , Zhou, H. , Xiang, H. , Zhu, Z. , Li, Y. , Lee, C. H. , and Xu, G. , 2015, “ CPS Modeling of CNC Machine Tool Work Processes Using an Instruction-Domain Based Approach,” Eng., 1(2), pp. 247–260. [CrossRef]
Mourtzis, D. , Doukas, M. , and Milas, N. , 2016, “ A Knowledge-Based Social Networking App for Collaborative Problem-Solving in Manufacturing,” Manuf. Lett., 10(1), pp. 1–5.
Mourtzis, D. , and Vlachou, E. , 2016, “ Cloud-Based Cyber-Physical Systems and Quality of Services,” TQM Emer. J., 28(5), pp. 704–733. [CrossRef]
Chui, M. , Löffler, M. , and Roberts, R. , 2010, “ The Internet of Things,” McKinsey Quarterly, Seattle, WA, accessed Sept. 28, 2017, http://www.mckinsey.com/industries/high-tech/our-insights/the-internet-of-things
Atzori, L. , Iera, A. , and Morabito, G. , 2010, “ The Internet of Things: A Survey,” Comput. Networks, 54(15), pp. 2787–2805. [CrossRef]
Perera, C. , Zaslavsky, A. , Christen, P. , and Georgakopoulos, D. , 2014, “ Context Aware Computing for the Internet of Things: A Survey,” IEEE Commun. Surv. Tutor., 16(1), pp. 414–454. [CrossRef]
Ding, F. , Song, A. , Tong, E. , and Li, J. , 2016, “ A Smart Gateway Architecture for Improving Efficiency of Home Network Applications,” J. Sens., 2016, p. 2197237.
Díaz, M. , Martín, C. , and Rubio, B. , 2016, “ State-of-the-Art, Challenges, and Open Issues in the Integration of Internet of Things and Cloud Computing,” J. Network Comput. Appl., 67, pp. 99–117. [CrossRef]
Luo, X. , Liu, J. , Zhang, D. , and Chang, X. , 2016, “ A Large-Scale Web QoS Prediction Scheme for the Industrial Internet of Things Based on a Kernel Machine Learning Algorithm,” Comput. Networks, 101(1), pp. 81–89. [CrossRef]
Lee, I. , and Lee, K. , 2015, “ The Internet of Things (IoT): Applications, Investments, and Challenges for Enterprises,” Bus. Horiz., 58(4), pp. 431–440. [CrossRef]
Wright, P. , 2014, “ Cyber-Physical Product Manufacturing,” Manuf. Lett., 2(2), pp. 49–53. [CrossRef]
Li, Y. , Thai, M. T. , and Wu, W. , 2008, Wireless Sensor Networks and Applications, Springer, New York. [CrossRef]
Ding, F. , and Song, A. , 2016, “ Development and Coverage Evaluation of ZigBee-Based Wireless Network Applications,” J. Sens., 2016, p. 2943974.
Teti, R. , Jemielniak, K. , O'Donnell, G. , and Dornfeld, D. , 2010, “ Advanced Monitoring of Machining Operations,” CIRP Ann.–Manuf. Tech., 59(2), pp. 717–739. [CrossRef]
Post, T. , Ilsen, R. , Hamann, B. , Hagen, H. , and Aurich, J. C. , 2017, “ User-Guided Visual Analysis of Cyber-Physical Production Systems,” ASME J. Comput. Inf. Sci. Eng., 17(2), p. 021005. [CrossRef]
Tapoglou, N. , Mehnen, J. , Vlachou, A. , Doukas, M. , Milas, N. , and Mourtzis, D. , 2015, “ Cloud Based Platform for Optimal Machining Parameter Selection Based on Function Blocks and Real Time Monitoring,” ASME J. Manuf. Sci. Eng., 137(4), p. 040909. [CrossRef]
Jardine, A. K. S. , Lin, D. , and Banjevic, D. , 2006, “ A Review on Machinery Diagnostics and Prognostics Implementing Condition-Based Maintenance,” Mech. Syst. Signal Process., 20(7), pp. 1483–1510. [CrossRef]
Si, X. S. , Wang, W. , Hu, C. H. , and Zhou, D. H. , 2011, “ Remaining Useful Life Estimation—A Review on the Statistical Data Driven Approaches,” Eur. J. Oper. Res., 213(1), pp. 1–14. [CrossRef]
Rehorn, A. G. , Jiang, J. , and Orban, P. E. , 2005, “ State-of-the-Art Methods and Results in Tool Condition Monitoring: A Review,” Int. J. Adv. Manuf. Tech., 26(7–8), pp. 693–710. [CrossRef]
Mourtzis, D. , Vlachou, E. , Doukas, M. , Kanakis, N. , Xanthopoulos, N. , and Koutoupes, A. , 2015, “ Cloud- Based Adaptive Shop-Floor Scheduling Considering Machine Tool Availability,” ASME Paper No. IMECE2015-53025.
Mourtzis, D. , Vlachou, E. , Xanthopoulos, N. , Givehchi, M. , and Wang, L. , 2016, “ Cloud-Based Adaptive Process Planning Considering Availability and Capabilities of Machine Tools,” J. Manuf. Sys., 39(1), pp. 1–8. [CrossRef]
Xu, X. , 2012, “ From Cloud Computing to Cloud Manufacturing,” Rob. Comput.-Integr Manuf.,” 28(1), pp. 75–86. [CrossRef]
Wu, D. , Thames, J. L. , Rosen, D. W. , and Schaefer, D. , 2013, “ Enhancing the Product Realization Process With Cloud-Based Design and Manufacturing Systems,” ASME J. Comput. Inf. Sci. Eng., 13(4), p. 041004. [CrossRef]
Wu, D. , Greer, M. J. , Rosen, D. W. , and Schaefer, D. , 2013, “ Cloud Manufacturing: Strategic Vision and State-of-the-Art,” J. Manuf. Sys., 32(4), pp. 564–579. [CrossRef]
Mourtzis, D. , Schoinochoritis, B. , and Vlachou, E. , 2015, “ A New Era of Web Collaboration: Cloud Computing and Its Applications in Manufacturing,” 8th IWC Total Quality Management Advanced and Intelligent Approaches Conference, Belgrade, Serbia, June 1–5, pp. 11–23. https://www.researchgate.net/publication/282094282_A_New_Era_of_Web_Collaboration_Cloud_Computing_and_its_Applications_in_Manufacturing
Li, W. , and Mehnen, J. , 2013, Cloud Manufacturing: Distributed Computing Technologies for Global and Sustainable Manufacturing, Springer, New York.
Wang, L. , Törngren, M. , and Onori, M. , 2015, “ Current Status and Advancement of Cyber-Physical Systems in Manufacturing,” J. Manuf. Sys., 37(1), pp. 517–527. [CrossRef]
Wang, Y. , and Nnaji, B. O. , 2005, “ Document-Driven Design for Distributed CAD Services in Service-Oriented Architecture,” ASME. J. Comput. Inf. Sci. Eng., 6(2), pp. 127–138. [CrossRef]
Erumban, A. A. , 2008, “ Lifetimes of Machinery and Equipment: Evidence From Dutch Manufacturing,” Rev. Income and Wealth, 54(2), pp. 237–268.
Mourtzis, D. , Vlachou, E. , and Milas, N. , 2016, “ Industrial Big Data as a Result of IoT Adoption in Manufacturing,” Proc. CIRP, 55(1), pp. 290–295. [CrossRef]
Tan, L. , and Jiang, J. , 2013, Digital Signal Processing, 2nd ed., Academic Press, London.
Gyftakis, K. N. , Spyropoulos, D. V. , Kappatou, J. C. , and Mitronikas, E. D. , 2014, “ Taking Advantage of the Induction Motor Inherent Eccentricity Aiming to Discriminate the Broken Bar Fault From Load Oscillations,” International Conference on Electrical Machines (ICEM), Berlin, Sept. 2–5, pp. 1933–1939.
Digi International, 2014, “ XBee/XBee-PRO ZB RF Modules, Product Manual,” Digi International Inc., Minneapolis, MN, accessed May 10, 2016, https://www.digi.com/resources/documentation/digidocs/PDFs/90000976.pdf
Mourtzis, D. , Vlachou, E. , Milas, N. , and Xanthopoulos, N. , 2015, “ A Cloud-Based Approach for Maintenance of Machine Tools and Equipment Based on Shop-Floor Monitoring,” Proc. CIRP, 41(1), pp. 655–660.
Mourtzis, D. , Vlachou, E. , Milas, N. , and Dimitrakopoulos, G. , 2016, “ Energy Consumption Estimation for Machining Processes Based on Real-Time Shop Floor Monitoring Via Wireless Sensor Networks,” Proc. CIRP, 57(1), pp. 637–642. [CrossRef]
ARM, 2009, “ Cortex-M4 Processor,” ARM Holdings Plc, Cambridge, UK, accessed Sept. 28, 2017, http://www.arm.com/products/processors/cortex-m/cortex-m4-processor.php
OPC Foundation, 2015, “ OPC Unified Architecture Specification—Part 5: Information Model,” OPC Foundation, Scottsdale, AZ, accessed Sept. 28, 2017, https://opcfoundation.org/developer-tools/specifications-unified-architecture/part-5-information-model
McKinsey & Company, 2015, “ Industry 4.0: How to Navigate Digitization of the Manufacturing Sector,” McKinsey & Company, New York, accessed Aug. 28, 2017, https://www.mckinsey.de/sites/mck_files/files/mck_industry_40_report.pdf
Copyright © 2018 by ASME
View article in PDF format.

References

Chryssolouris, G. , 2006, Manufacturing Systems: Theory and Practice, 2nd ed., Springer-Verlag, New York.
Mourtzis, D. , and Doukas, M. , 2014, “ The Evolution of Manufacturing Systems: From Craftsmanship to the Era of Customization,” Handbook of Research on Design and Management of Lean Production Systems, V. Modrák and P. Semančo , eds., IGI Global, Hershey, PA, pp. 1–29. [CrossRef]
Ilsen, R. , Meissner, H. , and Aurich, J. C. , 2017, “ Optimizing Energy Consumption in a Decentralized Manufacturing System,” ASME. J. Comput. Inf. Sci. Eng., 17(2), p. 021006. [CrossRef]
Löffler, M. , and Tschiesner, A. , 2013, “ Internet of Things and the Future of Manufacturing,” McKinsey & Company, New York, accessed Sept. 28, 2017, http://www.mckinsey.com/business-functions/business-technology/our-insights/the-internet-of-things-and-the-future-of-manufacturing
Feng, S. C. , Bernstein, W. Z. , Hedberg , T., Jr. , and Feeney, A. B. , 2017, “ Toward Knowledge Management for Smart Manufacturing,” ASME. J. Comput. Inf. Sci. Eng., 17(3), p. 031016. [CrossRef]
Bagheri, B. , Yang, S. , Kao, H. A. , and Lee, J. , 2015, “ Cyber-Physical Systems Architecture for Self-Aware Machines in Industry 4.0 Environment,” IFAC-PapersOnLine, 48(3), pp. 1622–1627. [CrossRef]
Sztipanovits, J. , and Ying, S. , 2013,“ Strategic R&D Opportunities for 21st Century, Cyber-Physical Systems—Connecting Computer and Information Systems With the Physical World,” National Institute of Standards and Technology (NIST), Gaithersburg, MD, accessed Sept. 28, 2017, http://www.nist.gov/el/upload/12-Cyber-Physical-Systems020113_final.pdf
Mikusz, M. , 2014, “ Towards an Understanding of Cyber-Physical Systems as Industrial Software-Product-Service Systems,” Proc. CIRP, 16(1), pp. 385–389. [CrossRef]
Monostori, L. , Kádár, B. , Bauernhansl, T. , Kondoh, S. , Kumara, S. , Reinhart, G. , Sauer, O. , Schuh, G. , Sihn, W. , and Ueda, K. , 2016, “ Cyber-Physical Systems in Manufacturing,” CIRP Ann.-Manuf. Technol., 65(2), pp. 621–641. [CrossRef]
Wang, L. , Shen, W. , and Lang, S. , 2004, “ Wise-ShopFloor: A Web-Based and Sensor-Driven e-Shop Floor,” ASME J. Comput. Inf. Sci. Eng., 4(1), pp. 56–60. [CrossRef]
Chen, D. , Maffei, A. , Ferreirar, J. , Akillioglu, H. , Khabazzi, M. R. , and Zhang, X. , 2015, “ A Virtual Environment for the Management and Development of Cyber-Physical Manufacturing Systems,” IFAC-PapersOnLine, 48(7), pp. 29–36. [CrossRef]
Chen, J. , Yang, J. , Zhou, H. , Xiang, H. , Zhu, Z. , Li, Y. , Lee, C. H. , and Xu, G. , 2015, “ CPS Modeling of CNC Machine Tool Work Processes Using an Instruction-Domain Based Approach,” Eng., 1(2), pp. 247–260. [CrossRef]
Mourtzis, D. , Doukas, M. , and Milas, N. , 2016, “ A Knowledge-Based Social Networking App for Collaborative Problem-Solving in Manufacturing,” Manuf. Lett., 10(1), pp. 1–5.
Mourtzis, D. , and Vlachou, E. , 2016, “ Cloud-Based Cyber-Physical Systems and Quality of Services,” TQM Emer. J., 28(5), pp. 704–733. [CrossRef]
Chui, M. , Löffler, M. , and Roberts, R. , 2010, “ The Internet of Things,” McKinsey Quarterly, Seattle, WA, accessed Sept. 28, 2017, http://www.mckinsey.com/industries/high-tech/our-insights/the-internet-of-things
Atzori, L. , Iera, A. , and Morabito, G. , 2010, “ The Internet of Things: A Survey,” Comput. Networks, 54(15), pp. 2787–2805. [CrossRef]
Perera, C. , Zaslavsky, A. , Christen, P. , and Georgakopoulos, D. , 2014, “ Context Aware Computing for the Internet of Things: A Survey,” IEEE Commun. Surv. Tutor., 16(1), pp. 414–454. [CrossRef]
Ding, F. , Song, A. , Tong, E. , and Li, J. , 2016, “ A Smart Gateway Architecture for Improving Efficiency of Home Network Applications,” J. Sens., 2016, p. 2197237.
Díaz, M. , Martín, C. , and Rubio, B. , 2016, “ State-of-the-Art, Challenges, and Open Issues in the Integration of Internet of Things and Cloud Computing,” J. Network Comput. Appl., 67, pp. 99–117. [CrossRef]
Luo, X. , Liu, J. , Zhang, D. , and Chang, X. , 2016, “ A Large-Scale Web QoS Prediction Scheme for the Industrial Internet of Things Based on a Kernel Machine Learning Algorithm,” Comput. Networks, 101(1), pp. 81–89. [CrossRef]
Lee, I. , and Lee, K. , 2015, “ The Internet of Things (IoT): Applications, Investments, and Challenges for Enterprises,” Bus. Horiz., 58(4), pp. 431–440. [CrossRef]
Wright, P. , 2014, “ Cyber-Physical Product Manufacturing,” Manuf. Lett., 2(2), pp. 49–53. [CrossRef]
Li, Y. , Thai, M. T. , and Wu, W. , 2008, Wireless Sensor Networks and Applications, Springer, New York. [CrossRef]
Ding, F. , and Song, A. , 2016, “ Development and Coverage Evaluation of ZigBee-Based Wireless Network Applications,” J. Sens., 2016, p. 2943974.
Teti, R. , Jemielniak, K. , O'Donnell, G. , and Dornfeld, D. , 2010, “ Advanced Monitoring of Machining Operations,” CIRP Ann.–Manuf. Tech., 59(2), pp. 717–739. [CrossRef]
Post, T. , Ilsen, R. , Hamann, B. , Hagen, H. , and Aurich, J. C. , 2017, “ User-Guided Visual Analysis of Cyber-Physical Production Systems,” ASME J. Comput. Inf. Sci. Eng., 17(2), p. 021005. [CrossRef]
Tapoglou, N. , Mehnen, J. , Vlachou, A. , Doukas, M. , Milas, N. , and Mourtzis, D. , 2015, “ Cloud Based Platform for Optimal Machining Parameter Selection Based on Function Blocks and Real Time Monitoring,” ASME J. Manuf. Sci. Eng., 137(4), p. 040909. [CrossRef]
Jardine, A. K. S. , Lin, D. , and Banjevic, D. , 2006, “ A Review on Machinery Diagnostics and Prognostics Implementing Condition-Based Maintenance,” Mech. Syst. Signal Process., 20(7), pp. 1483–1510. [CrossRef]
Si, X. S. , Wang, W. , Hu, C. H. , and Zhou, D. H. , 2011, “ Remaining Useful Life Estimation—A Review on the Statistical Data Driven Approaches,” Eur. J. Oper. Res., 213(1), pp. 1–14. [CrossRef]
Rehorn, A. G. , Jiang, J. , and Orban, P. E. , 2005, “ State-of-the-Art Methods and Results in Tool Condition Monitoring: A Review,” Int. J. Adv. Manuf. Tech., 26(7–8), pp. 693–710. [CrossRef]
Mourtzis, D. , Vlachou, E. , Doukas, M. , Kanakis, N. , Xanthopoulos, N. , and Koutoupes, A. , 2015, “ Cloud- Based Adaptive Shop-Floor Scheduling Considering Machine Tool Availability,” ASME Paper No. IMECE2015-53025.
Mourtzis, D. , Vlachou, E. , Xanthopoulos, N. , Givehchi, M. , and Wang, L. , 2016, “ Cloud-Based Adaptive Process Planning Considering Availability and Capabilities of Machine Tools,” J. Manuf. Sys., 39(1), pp. 1–8. [CrossRef]
Xu, X. , 2012, “ From Cloud Computing to Cloud Manufacturing,” Rob. Comput.-Integr Manuf.,” 28(1), pp. 75–86. [CrossRef]
Wu, D. , Thames, J. L. , Rosen, D. W. , and Schaefer, D. , 2013, “ Enhancing the Product Realization Process With Cloud-Based Design and Manufacturing Systems,” ASME J. Comput. Inf. Sci. Eng., 13(4), p. 041004. [CrossRef]
Wu, D. , Greer, M. J. , Rosen, D. W. , and Schaefer, D. , 2013, “ Cloud Manufacturing: Strategic Vision and State-of-the-Art,” J. Manuf. Sys., 32(4), pp. 564–579. [CrossRef]
Mourtzis, D. , Schoinochoritis, B. , and Vlachou, E. , 2015, “ A New Era of Web Collaboration: Cloud Computing and Its Applications in Manufacturing,” 8th IWC Total Quality Management Advanced and Intelligent Approaches Conference, Belgrade, Serbia, June 1–5, pp. 11–23. https://www.researchgate.net/publication/282094282_A_New_Era_of_Web_Collaboration_Cloud_Computing_and_its_Applications_in_Manufacturing
Li, W. , and Mehnen, J. , 2013, Cloud Manufacturing: Distributed Computing Technologies for Global and Sustainable Manufacturing, Springer, New York.
Wang, L. , Törngren, M. , and Onori, M. , 2015, “ Current Status and Advancement of Cyber-Physical Systems in Manufacturing,” J. Manuf. Sys., 37(1), pp. 517–527. [CrossRef]
Wang, Y. , and Nnaji, B. O. , 2005, “ Document-Driven Design for Distributed CAD Services in Service-Oriented Architecture,” ASME. J. Comput. Inf. Sci. Eng., 6(2), pp. 127–138. [CrossRef]
Erumban, A. A. , 2008, “ Lifetimes of Machinery and Equipment: Evidence From Dutch Manufacturing,” Rev. Income and Wealth, 54(2), pp. 237–268.
Mourtzis, D. , Vlachou, E. , and Milas, N. , 2016, “ Industrial Big Data as a Result of IoT Adoption in Manufacturing,” Proc. CIRP, 55(1), pp. 290–295. [CrossRef]
Tan, L. , and Jiang, J. , 2013, Digital Signal Processing, 2nd ed., Academic Press, London.
Gyftakis, K. N. , Spyropoulos, D. V. , Kappatou, J. C. , and Mitronikas, E. D. , 2014, “ Taking Advantage of the Induction Motor Inherent Eccentricity Aiming to Discriminate the Broken Bar Fault From Load Oscillations,” International Conference on Electrical Machines (ICEM), Berlin, Sept. 2–5, pp. 1933–1939.
Digi International, 2014, “ XBee/XBee-PRO ZB RF Modules, Product Manual,” Digi International Inc., Minneapolis, MN, accessed May 10, 2016, https://www.digi.com/resources/documentation/digidocs/PDFs/90000976.pdf
Mourtzis, D. , Vlachou, E. , Milas, N. , and Xanthopoulos, N. , 2015, “ A Cloud-Based Approach for Maintenance of Machine Tools and Equipment Based on Shop-Floor Monitoring,” Proc. CIRP, 41(1), pp. 655–660.
Mourtzis, D. , Vlachou, E. , Milas, N. , and Dimitrakopoulos, G. , 2016, “ Energy Consumption Estimation for Machining Processes Based on Real-Time Shop Floor Monitoring Via Wireless Sensor Networks,” Proc. CIRP, 57(1), pp. 637–642. [CrossRef]
ARM, 2009, “ Cortex-M4 Processor,” ARM Holdings Plc, Cambridge, UK, accessed Sept. 28, 2017, http://www.arm.com/products/processors/cortex-m/cortex-m4-processor.php
OPC Foundation, 2015, “ OPC Unified Architecture Specification—Part 5: Information Model,” OPC Foundation, Scottsdale, AZ, accessed Sept. 28, 2017, https://opcfoundation.org/developer-tools/specifications-unified-architecture/part-5-information-model
McKinsey & Company, 2015, “ Industry 4.0: How to Navigate Digitization of the Manufacturing Sector,” McKinsey & Company, New York, accessed Aug. 28, 2017, https://www.mckinsey.de/sites/mck_files/files/mck_industry_40_report.pdf

Figures

Grahic Jump Location
Fig. 1

The proposed IoT-based monitoring system

Grahic Jump Location
Fig. 2

The initiate communication protocol between one DAQ and the gateway

Grahic Jump Location
Fig. 3

Supervisory mechanisms to detect malfunctions in the network. In the left (a) is the flowchart of the DAQ, while in the right (b) is the flowchart of the Gateway.

Grahic Jump Location
Fig. 4

The block diagram of the developed DAQ

Grahic Jump Location
Fig. 5

The information model for the OPC-UA integration

Grahic Jump Location
Fig. 6

Measurements from a short-duration machining operation

Grahic Jump Location
Fig. 7

Measurements from a long-duration machining operation

Tables

Table Grahic Jump Location
Table 1 The list of selected PIs and their definition
Table Grahic Jump Location
Table 2 The task report that is generated for the long-term milling task
Table Grahic Jump Location
Table 3 The data bytes generated and transmitted to each level

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In