18-10-2016, 10:57 AM
1459573762-ApplicationsofIoTinproductionlogistics.rtf (Size: 513.05 KB / Downloads: 6)
Abstract
Companies and researchers have been pursuing Internet of Things (IoT) for dec-ades. With the rapid development of radio frequency identification (RFID) and cloud computing technologies, IoT has increasingly become technically feasible in many industries. As the earliest industry field related to IoT, production logistics will benefit from this innovative technology by improving the ability of tracing and tracking, responsiveness, resource management, decision support, etc. Meanwhile, challenges still exist in the implementation of IoT, e.g. integration of systems, in-formation security, Big Data analysis, and decision-making methods. In this paper, a brief introduction of the development of research on IoT in production logistics is provided as a reference for further research work.
Keywords: Internet of Things, radio frequency identification, production logistics.
1 Introduction
Internet of Things (IoT) was envisaged as a network of objects and sensors that collect and process information autonomously to enable computers to observe, identify, and gain situational awareness. The term IoT was linked to radio fre-quency identification (RFID) and supply chain management. After decades, the evolution of IoT has brought increasing opportunities and challenges in production logistics.
The collection, comprehension, and utilization of business data will affect the competitiveness of a company. Internet of Things solutions provide more con-nections among various systems and sensors via the broader Internet. Assets are digitally identified and automatically tracked with embedded technologies such as RFID.
WIT Transactions on Engineering Sciences, Vol 113, © 2016 WIT Press
www.witpress.com, ISSN 1743-3533 (on-line)
doi:10.2495/IWAMA150271
234 Advanced Manufacturing and Automation V
The rapid development of RFID technology sharply reduced the cost of identi-fying assets and has also made remote and unattended asset management possible. Combining with other sensors, RFID not only enhances the identification of mon-itored items, but also provides another way to identify the sensors themselves [1, 2]. However, there is not a single generally accepted IoT solution at present.
Production logistics encompasses logistic processes, which include all activities related to the supply of products with feedstock such as raw materials and work in progress (WIP), and the delivery of semifinished and finished products to the distri-bution center. Production logistics connects procurement logistics and distribution logistics. Thereby, the fundamental goal of production logistics can be formulated as the pursuance of greater delivery capability and reliability at the lowest pos-sible logistic and production costs [3]. However, in different logistic processes, the objectives are often competing
Since production logistics increasingly requires responsiveness, IoT techniques can contribute by facilitating real-time feedstock data collection, promoting a more collaborative software environment, and enhancing global and local decision-support capabilities.
2 Capability of IoT technologies
Compared with traditional machine-to-machine (M2M) technologies, IoT con-cerns more general Internet technologies. Communication among sensors, actuat-ors, and systems follows various network protocols, according to specific working conditions.
IoT technologies enhance data-acquisition capability. In a production process, raw data is collected, which is not meaningful until it is converted to information and, finally, knowledge for decision support. Information is defined as the enrich-ment of data with the right context, while data is perceived as raw values without
relevant or usable context [4]. The utilization of data comprises data visualiza-tion and data mining. The former focuses on the representation of data, combining with infrastructures or hardware systems (such as an item-tracking system) for monitoring or tracking assets, etc. [5, 6]. The latter focuses on the data mining process according to raw data and relevant context, which generates high-level information and even knowledge for decision support based on Big Data ana-lytics [7]. Generally, an IoT solution for production logistics is a cyber-physical system (CPS) (Fig. 1). On the basis of specific utilization, physical facilities dir-ectly related to production can be divided into three categories: manufacturing, transportation, and storage facilities. Manufacturing facilities include mainly ma-chines and robots used for processing feedstock. Transportation facilities refer to equipment that ship feedstock, such as pallets, conveyer belts, forklifts, automated guided vehicles, and robots. Storage facilities are designed for stocking feedstock and finished goods as a buffer in a warehouse.
3 Opportunities of IoT in production logistics
IoT can provide production logistics with a CPS enabled by Internet-based inform-ation communication technologies. Since the main advantage of IoT over M2M is the utilization of Internet techniques, the biggest contribution of IoT in produc-tion logistics will occur in the field where more connections can be built among hardware. Especially, RFID techniques can enable the tagging and automatic iden-tification of each equipment, device and even item [2], which constructs the IT infrastructure of an IoT system available for tracking and tracing [6, 8].
Opportunities for IoT in production logistics can be divided into the following: constructing a seamlessly integrated environment, enhancing the scalability, con-figurability, and extendibility of the production system, facilitating autonomous agent-based local collaboration, and promoting quick rescheduling and planning in production logistics.
3.1 Constructing a seamlessly integrated environment
The first contribution of IoT technologies is automated item registration. Histor-ically, automated identification and registration of items in the production logistic system have relied on barcode or QR code, which needs human operators or vision systems to scan the code and retrieve the identity of an item. Thereby, the regis-tration time of items in the production logistic system is a bottleneck in improving efficiency. An IoT system uses an RFID system as an alternative method for the registration of items in high-level management software systems such as an ERP system [9]. Without artificial interference, it is capable of achieving real-time item registration in the production logistic management system, e.g. working stations can be identified and items produced can be recorded automatically [6, 10–13].
3.2 Facilitating an autonomous agent-based local collaboration
On the basis of automated registration, an IoT system also provides the possibility of smart agent. Specifically, equipment shipped can be monitored and located in management systems [8, 14]. Furthermore, equipment can communicate with each other by following specific network protocols [15–17]. A distributed self-organized “community” of shipping equipment can be formed based on swarm intelligence activities.
3.3 Promoting quick rescheduling and planning in production logistics
Owing to improved communication capability and enhanced data collection in terms of both quantity and quality, the production logistic management system
will have more real-time data available for making production and logistic plans. Since decision support is vital for production logistics, this real-time data is of great value to make data-driven decisions for scheduling [18, 19] and resource allocation [20–22].
4 Challenges of IoT in production logistics
With the sharp increase in interest on IoT, researchers are keeping an eye on the challenges faced by IoT in applications, which originate from techniques, standards, security, research methodologies, etc. Considering RFID techniques as examples, the cost of one RFID tag is about a few cents, which is still more ex-pensive than printable bar code tags. Meanwhile, an RFID system needs readers and antennas as tag-interrogating devices, which will be a high initial investment for companies. Thus, a hybrid system mixing bar code technologies and RFID technologies is the most common solution currently.
Another challenge comes from standards of IoT, or in other words, a widely accepted standard for an IoT solution is not available as yet. Although standards (EPCIS, ALE, etc.) for RFID technologies are available, which are mainly de-signed for logistics and supply chain, additional efforts in the process of software system integration are necessary. This problem affects the extendibility of an IoT system.
System security is another challenge for IoT applications. An IoT system is In-ternet based, while databases and decision-support systems are generally based on remote “clouds.” Thus, collected data and communication systems must be protected from unauthorized access.
Decision-support systems, i.e. the processing of massive IoT data, pose the biggest challenge. An IoT system can produce much more data than an M2M system. However, it is nontrivial to make the data comprehensible and achieve decision support. Missing values are very common in practical systems, and will be more prevalent in IoT systems. Moreover, in a large-scale network, net-work latency can lead to asynchronous data, which should be avoided. Therefore, a missing-value-processing policy and a synchronization mechanism are often required [23].
Since an IoT solution generally comprises multiple domains, relationships between data have to be studied; ontology study can be considered a feasible method to study such relationships. Ontology study has been developed in sup-ply chain management [24]. The concept of semantic web has been proposed by the World Wide Web Consortium for sharing data and reusing across applications, enterprises, and communities [25]. Ontology research has already become another popular topic for IoT data analysis [26–29].
5 Conclusions
IoT is being increasingly accepted by companies in production logistics and other fields. Pioneers are implementing current IoT solutions in business while others
are following up. The benefits of management, optimization, and responsiveness derived from IoT solutions are tremendous, while the challenges of security, toler-ance, and efficient decision support cannot be ignored, areas where further research work will be carried out.