Abstract

This research project developed a framework for assessing digital transformation (DT) maturity in large government-owned corporations (GOCs), focusing on integrating building information modelling (BIM), digital engineering (DE) practices, value of information, data integration aspects, and critical information management competencies, relating to asset management, asset delivery, and asset performance. Recognising DT’s strategic importance in asset lifecycle management, researchers created a practical DT maturity assessment tool applied to an Australian GOC. The framework and Excel assessment tool outline DT maturity across seven categories and 56 indicators, using the analytical hierarchy process (AHP) to rank indicators based on expert input. The tool enables straightforward data input and DT maturity evaluation, helping organisations understand their current DT status and set improvement targets. Workshops with DT professionals validated the tool’s effectiveness through a case study, offering insights into DT maturity levels and a roadmap for technological integration and data-driven decision-making. The research highlights DT’s role in enhancing efficiency, innovation, and safety, emphasising the need for strategic alignment with organisational objectives.

Introduction

The building and infrastructure industries are gradually transitioning towards building information modelling (BIM) as an object-based digital representation of an asset’s physical and functional characteristics [1], [2] and digital engineering (DE) practices to enable more productive methods of planning, designing, constructing, operating, and maintaining assets, using digital processes [3]. As detailed later in the paper, “digital” encompasses approaches, tools and technologies such as BIM and DE together with data science, the Internet of Things (IoT), and AI solutions integrated to digitise assets throughout the lifecycle. Asset owners recognise the importance of digital transformation (DT) in delivering and managing assets [4], as it offers many benefits throughout the asset lifecycle and can enhance efficiency, value for money, productivity, innovation, and safety.

1.1. DT of large infrastructure asset ownersGovernments and government-owned corporations (GOCs) are typically owners and operators of large assets, such as buildings, infrastructure, and other asset holdings across the spectrum of services. Governments and industries must transition from ‘digital by exception’ towards ‘digital by default’ [5], [6]. The infrastructure sector should adopt an evidence-based approach to infrastructure planning, delivery, and operations, including BIM, DE, embedded sensors, and digital asset management [7]. DT offers significant potential value and benefits to government agencies that manage assets across the whole-of-asset lifecycle. DT can assist in strategic investing and partnering with local governments and industry to deliver more benefits, including [8]:

Conclusion

The dynamic evolution of the building and infrastructure sectors towards digitalised practices underscores the growing importance of asset owners engaging in comprehensive DT. This research project has developed a multifaceted framework for assessing DT maturity tailored to the needs of large GOCs.

The existing assessment tools and methods have been reviewed for suitability in assessing the DT in whole-of-life asset owner organisations. The reviewed tools offer varying assessment scopes, including a capability assessment, maturity assessment, and benchmarking. The need and potential benefit of a new maturity assessment tool that would balance assessment accuracy and completion efficiency were established based on the shortcomings and limitations of existing DT maturity tools and methods. The key contributions of the new DT maturity assessment framework and tool developed by the research team include (i) a holistic view on the assessment of how well organisations use data and digital solutions, focusing on BIM capabilities, the value of information, data integration aspects, components of the smart city concept and critical information management competencies; (ii) an evidence-based scoring approach including assigning weight to various assessment indicators based on their importance to ensure the accuracy of the assessment; (iii) a multiperspective assessment allowing for more accurate organisation-wide analysis; (iv) short-term and long-term recommendations based on the current and target levels gap to facilitate performance improvement while realising the benefits and value of getting to each maturity level.

Abstract

In the era of the competitive environment, the improvement in current products is ensured through activities aimed at increasing a product’s quality level and, consequently, reducing the amount of waste. The dynamically changing production environment and sudden changes in customer expectations force us to take precise and well-thought-out development steps. Furthermore, it is important to anticipate favourable product changes to prepare for market changes over time. This is still an open problem. The aim of this study was to develop a method to predict the quality of potential product prototypes resulting from the proposed modifications of the product features. This methodology takes into account current customer expectations. The method was created based on the principles of creating Quality Function Deployment (QFD) in the context of taking into account current and future customer expectations regarding product features. This is a new approach to analysing product quality within the principles of the traditional QFD method. The originality of the study is the technique used in the method to estimate the expected values of product features and their importance (weights), taking into account current customer expectations. Its originality is also manifested in drawing conclusions supporting the decision-making process of product improvement, because it involves ensuring the pro-quality modification of selected features of current products in the order that is most advantageous from the customer’s point of view. The use of the proposed method allows for the analysis of the impact of modifying the current value of a product feature. The method is illustrated with an example of a vacuum cleaner for home use. However, the proposed method can be applied to the design of any product to predict products that will meet customer expectations.

Introduction

The improvement in product quality covers a product’s customisation to expectations [1,2,3]. In this regard, it is necessary to have a skilful shape of the attributes of a product to achieve product quality that meets the level of expectation. This shaping includes acquiring the Voice of Customers (VoC) [4,5], and then involves its analysis and transformation in product attributes [6,7,8]. The product quality level mentioned is the level of compliance of a product with customers’ expectations and refers only to specific product uses. On the other hand, customer expectations (i.e., their desires and assumptions in the context of the future) are customer satisfaction with what they received (current product) and what they expected (product they expected) [9,10,11]. Despite that, designing products is still a problem; therefore, a search for methods to solve this problem is needed.

After reviewing the literature, it was concluded that the House of Quality (HoQ) [12] i.e., Quality Function Deployment (QFD) [13], is the most popular and most-used method to design a satisfactory product [14,15,16,17,18]. This method is a correlation matrix. The method supports the process of design by collecting and verifying expectations, e.g., of customers and the producer, to achieve the required product quality level. Previous work as part of HoQ concerns mainly the validity (weights) of customer requirements, e.g., mainly as part of the use of the AHP method (Analytic Hierarchy Process) [15,19,20,21]. Also, as part of HoQ, the customers’ requirements were specified, mainly using the fuzzy Saaty scale and FAHP method (Fuzzy Analytic Hierarchy Process) [16,22,23]. For this purpose, the HoQ with the AHP method and the Kano model have been integrated relatively often [21,24,25], where the objective of this combination is to precisely determine the importance of customer expectations in the context of product satisfaction. This precision was achieved by using the 2-tuple linguistic [17], in which the customer’s needs were characterised by description and number, which is a distance of the central value of the linguistic term. Despite this, the importance of customers’ requirements in the HoQ was determined by implementing a mathematical model according to LGP (Linear Goal Programming), in which the ordinal scale was used [26]. Another example is a combination of the HoQ with the Yager algorithm [27] to determine customer requirement weights also in the case of indifferent requirements. Another example concerning the design of the expected product is the combination of the HoQ method with the TRIZ method (Theory of Innovative Problem Solving) [28,29,30]. This combination encompasses determining customers’ requirements and their expectations as part of providing a satisfactory product.

Conclusions

In a competitive environment, it is reasonable to look for various solutions to improving products. This mainly concerns shaping the current features of products as part of customising them to customers’ expectations. However, designing a satisfactory product ahead of the competition is still a problem. This refers to the prediction of the quality level of the expected product, which can be designed based on the current customer expectations. For this purpose, a method was proposed to predict the pro-quality modification of product attributes considering current customers’ expectations. The method was developed by transforming the House of Quality in the context of considering the values of the attributes of the current and expected product.

The proposed method was tested by using a domestic vacuum cleaner as an example. The purpose was to predict the quality of the modification of the attributes of the vacuum cleaner considering the expectations of current customers. By using a survey and the Likert scale, the requirements of 166 customers were obtained. In the research, 20 attributes of the vacuum cleaner were included, for which the customers assessed their satisfaction with the current and modified values of the features. In addition, customers determined the importance (weights) of the product attributes. According to the Pareto rule, the important attributes were selected. The important product attributes were the vacuum in the suction pipe, working range of the vacuum cleaner connected to the power cord, length of the power cord, and vacuum cleaner motor power. For these attributes, the current and favourable modification values were determined. As part of the calculation model, the quality of the product attributes was estimated. In addition, the importance and influence of the modification of the current product attribute values were estimated. As a result, the pro-quality product modifications were predicted. Therefore, a classification (order) of the changes of the product attributes in the context of achieving the expected level of product quality was predicted. It has been predicted that in the first order, the working range (above 19 m) and motor power (above 900 W) should be changed. The change in these attributes allows for an expected increase in the product quality level. The next anticipated changes were length of the power cord to above 15 m and then vacuum in the suction pipe to above 27,000 Pa. Herein, the manufacturer has the final decision as to which of the product attribute modifications will be implemented.

Abstract

     The literature on quality-management standards has generally focused on the drivers, motivations, and performance effects of adopting such standards. Yet the last decade has witnessed a substantial degree of decertification behavior, as organizations have increasingly decided to voluntarily withdraw from quality-management standards by not recertifying. While the drivers of the decision to initially adopt quality-management standards have been extensively studied, the drivers of the decision to decertify have received scant scholarly attention. We argue that innovative organizations are generally prone to retaining quality-management certification and thus exhibit a tendency to not abandon certification; however, radically-innovative organizations are more prone than incrementally-innovative organizations to discontinue quality-management standards and thereby exhibit a tendency to withdraw from quality certification. We compile World Bank data surveying facilities based in 50 countries and 103 industrial sectors across the 2003 to 2017 period. Taking advantage of the data’s panel properties yields a dataset composed of up to 1755 facility-level observations of recertification decisions for empirical analysis. Our empirical testing employs a probit estimation technique that accounts for the appropriate fixed effects and generates results that support our theoretical priors regarding decertification behavior.

Introduction

     Certification in internationally-recognized quality standards (e.g., ISO 9000, QS 9000, ISO 13485, and IATF 16949) across the globe by millions of firms and facilities over the last three decades has spurred a great deal of literature—see the reviews by Corbett and Yeung (2008), Heras-Saizarbitoria and Boiral (2013), and Castka and Corbett (2015). While many studies have focused on the internal and external performance effects of adopting these standards (e.g., Blind, 2001; Hendricks and Singhal, 2001; Corbett et al., 2005; Martínez-Costa et al., 2009; Levine and Toffel, 2010; Singh et al., 2011), Anderson et al.’s (1999) pioneering study established that deciphering the forces which lead to the adoption of quality-management standards represents a central research question within this literature. Factors such as organizational size, firm age, government mandates, customer pressure, supply chains, and export considerations (Corbett and Kirsch, 2001; Vastag, 2004; Corbett, 2006; Delmas and Montiel, 2009) have all been considered instrumental in explaining the decision of a particular facility to incur the costs to seek and obtain a quality-management standard.

Discussion and conclusions

     While a healthy amount of scholarship exists regarding the factors influencing organizational adoption of quality-management standards, Castka and Corbett (2015), Cândido and Ferreira (2021b) and others have pointed out the share scarcity of scholarship focusing on the decision to voluntarily withdraw and decertify from quality standards. This relative neglect with respect to the drivers of decertification is somewhat out of step with a contemporary reality where organizations have been increasingly abandoning quality standards over the last decade (Castka and Corbett, 2015; Cândido et al., 2016, Cândido et al., 2021; Kafel and Simon, 2017; Mastrogiacomo et al., 2021). By neglecting the potential for withdrawal from quality-management standards, the certification literature has made incorrect inferences regarding the universal presence of continued adoption behavior. That is, the quality-certification literature has essentially assumed that adoption pressures unequivocally build for a focal organization when in fact these pressures can be alleviated to such a degree that organizations will begin to withdraw from quality standards.

Abstract

1. Introduction

2. Background and review of literature

3. Discussion

4. Conclusion and Future Scope

Acknowledgment

References

Abstract

     Enterprise Resource Planning (ERP) system is a collection of collaborative software programs. It handles transactions through enterprise-wide business processes using shared databases, standard methodologies, and data exchange across and within functional domains. Setting up an enterprise system is a complex activity and a costly and dangerous investment. Further, ERP system potentially affects core business and supporting processes, especially in complex and cyber-physical domains such as Industrial Internet of Things (IIoT) and Smart Factory. Cloud ERP (C-ERP) and Edge ERP (E-ERP) are alternatives to traditional, centralised and monolithic ERP implementation for incorporating the benefits of Cloud and Edge Computing. Their main benefits include ease of use, resource balancing, bandwidth, cost-saving, and higher privacy/security. This paper discusses the benefits and limitations of using C-ERP and E- ERP in IIoT and Smart Factory domains, along with future directions in the ERP era of demand.

Introduction

     One of the most widely utilised information technology (IT) solutions in organisations are ERP systems [1]. These are well-organised collections of software that works together to timely integrate business processes and support the management of effective cross-functional operations within an organisation. In particular, in ERP systems, a single database has all of the information. It acts as a central location for storing, distributing, and sending data between departments and business processes to manage transactions and data exchange across and within different functional areas. Setting up such a complex, shared and monolith system as an ERP is more than a computer project. It’s a costly and hazardous investment, and the process impacts the organisation’s core and support operations [2]. However, business operations’ technical and functional integration to synchronise the flow of information with the content flow of products or services is very attractive for the companies, representing one of the significant reasons for ERP adoption [3] [4].

     Simultaneously, companies can also benefit from novel enabling technologies such as Cloud and Edge computing. First, they provide the capability to process and store large amounts of data. Second, they allow continuous and realtime information sharing between customers and consumers; and third, they can rapidly and agilely respond to consumer behaviour and organisation needs [5]. The main benefits and features of Cloud and Edge computing perfectly match with ERP systems’ needs. Indeed, Cloud-based ERP (C-ERP) and Edge-based ERP (E-ERP) systems consist of a host ERP that operates, respectively, on a cloud and edge provider’s infrastructure. These infrastructures include software, hardware, storage, networks and other resources which can overcome the limits of conventional, centralised ERP systems. C-ERP and E-ERP solutions have been recently developed for many application domains, including the Industrial Internet of Things (IIoT) and Smart Factories. ERP systems can simplify the facility and business automation and support the integration of internal/ external resources and management decisions. Indeed, especially in such domains, bringing data and cyber-physical processes together by simultaneously complying with business and organisational needs is key for addressing the increasingly complex challenges of economy 4.0.

Conclusion and Future Scope

     The adoption of C-ERP and E-ERP is a strategic choice made by any industry. ERP’s competitive advantage stems from its integration with newer technologies (IIoT), which hides the constant need for change in an organisation’s operation. The study of the industrial decision-making process began with examining the factors that impact C-ERP and E-ERP adoption. E-ERP solutions can beat C-ERP solutions in terms of faster provisional delivery, better query execution, and closeness to the client. Edge computing may also be utilised in a variety of ERP systems and modules. Edge nodes may handle specific task allocations right at the edge network, simplifying and dividing activities to be done partially at the edge and transmitting only the most intensive computing jobs to C-ERP. Finally, the E-ERP model delivers real-time data, making it a more effective and efficient decision-making and reporting tool. Sharing workloads with the central ERP application server and controlling partial execution at the edge nodes achieve low latency and high performance.

Abstract

     Considering constantly increasing global competition in the market and developing technologies, information systems (ISs) have become an important component of the business world and a vital component of intelligent systems. An IS provides support for planning, controlling, analyzing activities, and support in decisions by managing data throughout the organization to assist executives in their decisions. The main function of an IS is to collect data spread between various parts of the organization and business partners and to process these collected data to form reliable information, which is required for decision making. Another critical function of an IS is to transfer the necessary information to the point-of-need in a timely manner. ISs assist in the conversion of data and information into meaningful outcomes. An IS is a combination of software, data storage hardware, related infrastructure, and people in the organization that use the system. Many business organizations rely on management information systems (MISs), and they conduct their critical operations based on these systems. The existence of an efficient MIS is a requirement for the sustainability of any business. However, MIS’s efficiency depends on the business’s requirements and nature. The compatibility of MIS with business in the company is vital for the successful implementation of these systems. The current study analyzes differences in expectations of manufacturing and service industries from MISs. For this aim, a fuzzy multi-criteria group decision-making (F-MCGDM) model is proposed to determine the differentiating success factors of MIS in both manufacturing and service industries. Findings indicate that there are considerable differences in the needs of both industries from MIS.

Introduction

     The availability of processed data is critical for any type of business, especially in a highly competitive environment, because real-time access to correct information provides executives of an organization to achieve an advantage against their competitors by taking efficient, effective, timely, and strategically reliable decisions. Collecting the required data, processing to convert it into information, and storing this information are the main functions of an IS. These systems are defined as the combination of people, hardware, software, data, procedures, and policies [1]. People in the system are called end users or clients, who use, develop and operate the system. IS design is made toward end users’ requirements by system analysts. Hardware includes computers and other equipment related to the process, storage, and transfer of data. Additionally, this hardware is a physical interface between the IS software and the end user. All these components interact with each other to handle information that flows inside the organization, and in many cases between peers in the supply chain.

     Most organizations require an information technology (IT) infrastructure. IT is concerned with modern enterprise management functions such as manufacturing, marketing, banking, and financing [2]. A well-designed IT infrastructure is a necessity for a well-performing IS. If IS is thought of as a production process in a manufacturing facility, information can be assumed as the final product. If the final product (information) cannot be delivered under a certain lead time and if it does not meet the customers’ (clients’) requirements, customers will not be satisfied with the product, and the company will lose its customers and future revenues [3]. IT contributes to innovation efforts by supplying increased communication and coordination between different departments and helps reduce the cost of transactions in an organization [4]. A successfully implemented IS, supported by a sufficient IT infrastructure, increases the performance of the whole supply chain of the organization via increased flexibility achieved in all operations between the peers in the chain [5].

Conclusions

     Requirements of today’s businesses increased the importance of ISs. Rapidly changing market conditions and customer needs force companies to react quickly to these changes. Adapting to constantly developing technological trends requires the existence of an effective MIS in organizations to aid executives in their decisions to acquire a sustainable business. However, ensuring the successful implementation of MIS requires a good fit of these systems with the organization’s business. Understanding the differences in business needs and differentiating factors related to MIS’s success must be determined for different industries. Therefore, an F-MCGDM framework is proposed in this study to fill the gap in the literature by determining the differences in expectations of manufacturing and service industries from MIS.

     The application of the proposed F-MCGDM framework to manufacturing and service industries to evaluate the success factors of MISs revealed a considerable difference between the expectations and requirements of companies in these industries from the MISs. Selecting the right MIS according to the most important factors of the company by considering the differentiating factors is important for the actual performance of MISs after the implementation.

1- Introduction

2- Methodology

3- Application of the model to design of RAF

4- Results and discussion

5- Conclusion

References

Abstract

In this article, a novel hybridized Multi-Attribute Decision Model (MADM) is developed to identify an optimal design of a Reconfigurable Assembly Fixture (RAF) from a set of alternative design concepts. The model combines the comparative advantage of Fuzzy Analytic Hierarchy Process (FAHP) and the computational strength of the Fuzzy Weighted Average (FWA) based on left and right scores in order to obtain aggregates for the design alternatives considering the relative importance of the design criteria as needed in the optimal design. The model was applied to evaluate four design concepts of a RAF with six design features having numerous sub-features. Results obtained from the evaluation process shows that there are differences in final values of the design alternatives. However, a close variation exists between these values. These differences can be accrued to the interrelationships between the design features and sub-features obtained from the Fuzzy Synthetic Extent (FSE) of the FAHP and an unambiguity judgment of the FWA when aggregating availability of the design features and sub-features in the design alternatives.

Introduction

Robust design of Products and industrial machineries is important from conceptualization to manufacturing and up till usage in order for manufacturers to obtain a share of the competitive market flooded with changeable designs (Olabanji, 2018). The need for these robust designs calls for development of different design concepts of a particular product or machine before a detail design analysis can be done (Song et al., 2013). The availability of alternative design concepts necessitates the need for selection of optimal design concept (Wei and Chang, 2008). Concept selection in engineering design has attracted importance in recent times because it has a direct implication on the quality of the final design. Problems that arises when it is not properly done includes; repetitive alterations and modifications of drafted designs, prolonged developmental time and amplified cost of actualization (Tiwari et al., 2017). In order to arrive at robust design of a new product or engineering component, identification of design attributes and sub features viz a viz the numerous functional requirements from the customers or intended end users becomes the first task (Ayag and Ozdem, 2007; Brackea et al., 2017). The engineering design process attempts to give a holistic approach to identification of the design attributes, sub features and functional requirements. It follows an established design standard by proposing four phases (product planning and clarification of task, conceptual design, embodiment design and detail design). These phases are usually applied to arrive at a detail design of the new product (Yeo et al., 2004; Olabanji and Mpofu, 2014). Also, the engineering design process can be imagined to have a set of eleven steps as described by (Ayag and Ozdem, 2007). The relationship between these two analogies is described in Figure 1 where these four phases are disintegrated into the eleven steps. It may be assumed that the information needed in each step and phase will also follow the same manner. However, since the concept selection step is a decision-making process, adequate information is needed for successful selection process. Concept selection in engineering design can be modelled as a multicriteria decision-making (MCDM) problem since it involves multiple design attributes that are having different sub features. Considering the steps in the engineering design process as an all-inclusive approach, it is possible to develop a relationship between the constraints, design attributes and sub features for determining optimal design concept using the multi-criteria analysis as presented in Figure 2.

1- Introduction

2- Research methodology

3- Technologies behind blockchain

4- Blockchain architecture

5- Salient features of blockchain

6- Drivers, barriers and risks in blockchain

7- Applications of blockchain

8- Conclusions

References

Abstract

The dawn of the 21st century has seen the advent of many technologies targeting commercial and financial sectors. These include Big Data, Internet of Things and FinTechs such as blockchain. Blockchain is a type of a distributed database that is used to replicate, share, and synchronise data spread across different geographical locations such as multiple sites, countries, or organisations. The main property of blockchain is that there is no central administrator or centralised data storage mechanism. Consensus algorithms govern the peerto-peer decentralised network. Numerous benefits and applications of blockchains have resulted in it becoming popular among a broad spectrum of businesses, but is it the case in the construction industry? Given, the backward nature of the construction industry in digitalisation and its reticence to change it becomes important to analyse the potential impact of Blockchains as a potential disruptive technology. Although there exists a significant research gap and the potential possibility to test blockchain in the construction sector, the construction industry is historically reported as the second lowest sector to have adopted information technology. This leads to a conundrum whether blockchain is a pure technological hype or whether there is a real potential application in construction. The paper is aimed at critically analysing the application potential of blockchains in construction through a use case analysis and comprehensive literature review to resolve whether it is pure hype or real. The exploration revealed that due to the exponential uses of blockchain, investments involved, and a number of start-up businesses contributing to Industry 4.0, blockchain indeed has a credible potential in the construction industry.

Introduction

Blockchain is considered a potential disrupter of the status core in the commercial sector innovating in transactions, revolutionising industries and driving economic change on a global scale. Blockchain technology is predicted to completely overhaul the digital economy [1]. Blockchains attracted attention from various industries such as finance, insurance, logistics, energy and transportation in experimenting its applications [2]. Though blockchain is a potent disruptor in different industries, it is important to understand whether this is just hype or real when applied to real-life problems. This paper is aimed at critically analysing whether blockchain is pure hype or whether there is credible application potential in the construction industry. The 4th industrial revolution (Industry 4.0), follows Industry 1.0 (water and steam power), Industry 2.0 (electricity) and Industry 3.0 (internet, electronic devices). It is characterised by the seamless integration of cyber-physical environments propelled by an array of technologies that enable the development of a digital and automated industry as well as the digitisation of the value chain [3-5]. Blockchain follows the egalitarian ideologies of the modern society where equality, direct dealing, openness, consensus and mutual trust play a vital role. Sikorski, et al. [6] states that blockchain technology has the potential to revolutionise the engineering industry by changing its production and procurement methodologies. A recent global report by McKinsey ranked the construction industry as the second lowest sector to have adopted information technology during Industry 3.0 [7]. Hence, there lies the conundrum, whether blockchain will create just hype or real disruption in construction, similar to the digital revolution that is taking hold in other industries.

The industrial applications in the cloud do not meet the requirements of low latency and reliability since variables must be continuously monitored. For this reason, industrial internet of things (IIoT) is a challenge for the current infrastructure because it generates a large amount of data making cloud computing reach the edge and become fog computing (FC). FC can be considered as a new component of Industry 4.0, which aims to solve the problem of big data, reduce energy consumption in industrial sensor networks, improve the security, processing and storage real-time data. It is a promising growing paradigm that offers new opportunities and challenges, beside the ones inherited from cloud computing, which requires a new heterogeneous architecture to improve the network capacity for delivering edge services, that is, providing computing resources closer to the end user. The purpose of this research is to show a systematic review of the most recent studies about the architecture, security, latency, and energy consumption that FC presents at industrial level and thus provide an overview of the current characteristics and challenges of this new technology.

Introduction

The fast development of technology has contributed to the increase of devices connected to the cloud, which generates large amounts of data. The cloud is the biggest data unit where processing and storage is performed with the main goal of making use of services and resources sought by the customers [1]. In 2015, the number of connected devices was 15.41 billion, later in 2017 it was of 20.35 billion, and it is expected to be 30.73 billion in 2020 [2]. These services and applications are used in different scenarios in the world such as Smart Factory, Smart Farming and Smart Cities [3]. Due to this rapid increase of data, a large amount of storage is required, which has generated a greater bandwidth consumption and high latency in data processing. The development of portable computing processes, more intelligent measurements in homes/cities and vehicles, and wider wireless sensor networks has made everything interconnected through the IoT [4]. Due to the number of interconnected devices, technology is generating big quantity of data that are processed, filtered and analyzed in the cloud causing problems related with traffic congestion, delays, and privacy concerns [2, 5]. Many limitations also emerge since the cloud is unable to support some of the needful such as heterogeneous devices, low latency, mobility, and location recognition [6]. In addition, the current infrastructure is not designed for the quantity and speed of data generated by IoT [7], which has generated a high consumption of resources and a decrease in the quality of services for the end user.

1. Introduction

2. Research Background and Motivation

3. Open Source Machine Learning Frameworks and Implementation Process

4. Results and Discussion

5. Conclusion

References

Information and communication technology has revolutionized the industrial operations and productions. The industries irrespective of size, whether small or large, have felt the need of artificial intelligence and machine learning techniques to process the terabytes of data generated through sensors, actuators, industrial management systems, and web applications. These data have the characteristics of volume (terabyte) and variety (image, audio, video, graphics) and thus customized models and techniques are required for analysis and management. The advancement in computer hardware, processing power, storage capacity, and cloud computing have led to experimentation and implementation of machine learning models in industrial domain for resource optimization, operation management, and quality control. However, the industrial Data Analysts face the dilemma of selecting the affordable and easy to use machine learning frameworks that suite their need and expectations. The study investigates the open source machine learning frameworks, aligned with the industrial domain (processing data generated from Industrial Internet of Things), in terms of usage, programming languages, implementations, and future prospects.

Introduction

Machine learning applications are quickly transforming the industrial landscape. Many businesses have reduced the production and operation costs using tools powered by machine learning models and algorithms. The deep learning which is a subset of machine learning has found ways in manufacturing, industrial maintenance, drug discovery, pattern imaging analytics, and software testing [1]. The deep learning a type of deep neural network consisting of layered structure as input layer, hidden layer, and output layer. Industrial Internet of Things (IIoT) is defined as a set of machines, robotics, cognitive technologies, and computers for intelligent industrial operations with the help of data analytics [2]. The Industrial Internet of Things is a part of Industry 4.0 revolution, which is concerned with automation, innovation, big data, and cyber physical systems in industries. The Industrial Internet of Things are showing positive impact in supply chain, transportation, healthcare, manufacturing, oil and gas, energy/utilities, chemical, and aviation industry. The Industrial Internet of Things has helped in controlling and monitoring manufacturing and production from remote locations [3-5]. The Industrial Internet of Things market will reach $123.89 Billion by 2021 [6]. Industrial Internet of Things captures large chunk of data, later used for predictive maintenance, time management, and cost control after machine learning models implementation. The machine learning models forms the core of logistics and supply chain solutions in terms of optimizing the product packet size, delivery vehicle selection, delivery route selection, delivery time computation. For instance DHL uses Amazon’s Kiva robotics (improve speed, accuracy) for the network management.

1- Introduction

2- System overview

3- Methodology

4- Experiments

5- Transfer learning

6- Conclusion

References

Abstract

Machine vision technologies have been widely used for automating the product quality control, but the defect inspection for codes on complex backgrounds is still a challenging task in the plastic container industry. In this work, an efficient and accurate inspection solution based on deep learning was proposed aiming at the detection of codes on complex backgrounds for the plastic container such as beverage packages. Firstly, image processing algorithms such as the region translation method, morphological processing, and image matching technology based on SIFT (Scale Invariant Feature Transform) features were implemented to generate synthetic defective samples, which moderated the class-imbalance problem. Data augmentation strategies were used to increase the amount of training data. Secondly, the ShuffleNet V2 framework was adapted to inspect inkjet codes on complex backgrounds. Additionally, the transfer learning was used to transfer the trained model to other inspection tasks for different kinds of packages. Finally, the proposed approach was built onto an in-line code inspection apparatus for the plastic container industry, and an accuracy of 0.9988 was achieved. The in-line testing results of false detection and omission detection rates demonstrated that the proposed solution can fully meet the production requirements. To the best of our knowledge, this report describes the first time that deep learning has been applied to the industrial defect inspection for the plastic container industry.

Introduction

The plastic packaging technologies have been experiencing explosive growth in the beverage industry. In order to ensure the product traceability and quality control, batch number and data codes are marked on the curved plastic surfaces early in the production process. The inkjet printing process, as the most commonly used type of printer, has been widely adopted to spray a digital image of the expiration date and manufacturing date by propelling droplets of ink onto the container. The quality of characters is affected by the performance of code printer and other external factors, and Legible codes on the containers provide consumers with important information and confidence for the inside product. If containers or packages with defective codes such as missing, incorrect, or unreadable codes are not identified in time, the product quality and corporate reputation will be compromised. Therefore, the code inspection that verifies the presence, position, and formation of printed codes is a key checkpoint in the plastic container industry to ensure products meet specifications prior to release and shipment to customers. With the development of machine learning and image processing technologies [1], the Optical Character Recognition (OCR) technology has been widely used by companies in the field of automatic code inspection on product containers, which greatly improves the accuracy and efficiency of code detection, and reduces their production costs as well as increases their profits. The general code detection is a composite process that comprises several phases such as the preprocessing, segmentation, feature extraction, and classification. The principal purpose is to compare the recognized characters with the correct codes and judge whether they are qualified. The widely used pixel binarization method for the character segmentation is the threshold-based segmentation method, and characters are segmented by setting a fixed or dynamic threshold.

Introduction

Review of relevant literature

Case description of predictive maintenance

Conclusions

References

Abstract

The Industry 4.0 era requires new quality management systems due to the ever increasing complexity of the global business environment and the advent of advanced digital technologies. This study presents new ideas for predictive quality management based on an extensive review of the literature on quality management and five realworld cases of predictive quality management based on new technologies. The results of the study indicate that advanced technology enabled predictive maintenance can be applied in various industries by leveraging big data analytics, smart sensors, artificial intelligence (AI), and platform construction. Such predictive quality management systems can become living ecosystems that can perform cause-effect analysis, big data monitoring and analytics, and effective decision-making in real time. This study proposes several practical implications for actual design and implementation of effective predictive quality management systems in the Industry 4.0 era. However, the living predictive quality management ecosystem should be the product of the organizational culture that nurtures collaborative efforts of all stakeholders, sharing of information, and co-creation of shared goals.

Introduction

In today’s competitive global environment, businesses need to be agile, flexible, resilient, and possess dynamic capabilities [1, 2]. The advent of advanced digital technologies makes it possible for firms to completely innovate the concept of quality management. A living ecosystem equipped with advanced digital technologies (e.g., smart sensors, machine learning, big data analytics, and artificial intelligence (AI)) can be developed to manage quality [2]. On August 14, 2018, a 200-m section of the Ponte Morandi Bridge (built in 1968) in Genoa, Italy, collapsed causing 41 deaths, 5 missing, and 15 injured. The main causes of bridge collapse were aging and lack of bridge management. Incidents such as this highlight the importance of bridge maintenance. Structural health monitoring (SHM), a new technique developed for structure maintenance, is an up-to-date technology-based system that analyzes weaknesses of existing systems, such as locating local and global damage structure and the significance of such damages. The speed and precision of decision-making for bridge repair and maintenance are facilitated by real-time monitoring of bridge conditions. Bansal et al. [3] proposed a real-time predictive maintenance system using neural network methods, while Shi and Zeng [4] suggested a condition-based maintenance strategy that considers economic factors for predictive maintenance in real time. Predictive maintenance, also known as condition-based maintenance, is possible today due to the advanced digital technologies [3–6].