Machine learning applied to quality management—A study in ship repair domain
Introduction
Ship repair is a complex, highly dynamic and stochastic process with high interdependencies. The process is also characterised with a high share of incomplete and unreliable information that is particularly expressed in some stages of the process. In such processes output quality is significantly influenced by the quality of assessments and decisions that cannot be ensured only by adherence to certain predefined procedures and instructions, on which, e.g. the standard ISO 9001 is based. In such systems expert knowledge and experience play a decisive role, and they are often of the nature that makes it practically impossible for them to be formalised with traditional methods. Also, because of so expressed technological complexity, and too many inter and intra dependent variables of influence, it is not easy (or even possible) to define efficient analytical models. Delivery time estimate in ship repair is one of typical examples of such processes. It includes the overall repair time estimate, as well as the estimate of duration of repair works in dock. The accuracy of these estimates significantly influences the quality of ship repair service. Also, it is critical for the business results of the shipyard. If the estimated times are too long, the shipyard will not be competitive. And if they are estimated too short, a production schedule may fail due to unrealistically estimated activity durations, which may result in final delivery time delay and penalties. Also, the quality of performed job might be influenced negatively given that delay often means doing things in hurry. This particularly goes for the overall repair time estimates.
On the other hand, developments in artificial intelligence provide powerful means for modelling expert knowledge. They also allow the automatic acquisition of such knowledge by means of machine learning or data mining techniques. Unfortunately, the use of such techniques in quality management context is not of systematic but rather of an ad hoc manner. In industry this is caused by at least two main reasons. The first is Taylorian philosophy of manufacturing that still prevails in the current quality management models. Determinism of operations, predictable behaviour of the system, and a priori information that is reliable, complete and accurate, identified as the basic Taylorian presumptions of manufacturing systems by Peklenik [1], are still the main presumptions of the most well known quality management models (total quality management model (TQM), Malcolm Baldrige Criteria for Performance Excellence, EFQM Excellence Model, and standard ISO 9001). For example, fact-based management, i.e. the factual approach to decision making, are still listed among core quality concepts in the frame of all these models. Also, the use of information technology is not sufficiently systematic. One of the consequences of this is the lack of accurate and standardised bases of organisational as well as of technological data in some manufacturing organisations and domains. The second reason why the use of artificial intelligence techniques in quality management context is not of systematic but rather of an ad hoc manner is that knowledge of artificial intelligence techniques is typically modest. On the other hand, although the Malcolm Baldrige criteria included recently knowledge management into one of its categories, the emphasis in related research is mostly on learning, i.e. on knowledge creation and knowledge sharing, and not knowledge formalisation process (see, e.g. [2]). Also, distinction between the terms ‘knowledge’ and ‘information’ is not always clear in such research (see, e.g. [3]). A more detailed explanation of these limitations, as well as the DQC model—a new theoretical framework how to overcome these deficiencies are presented by Srdoc et al. [4]. In difference to other quality models that are typically concerned only with shallow knowledge, in this model particular attention is paid to standardisation of domain concepts, and domain deep knowledge. Integration of information systems, defined as systems whose purpose is to acquire and represent knowledge, and quality systems is also proposed in [5]. Dooley [6] also suggests that TQM paradigm based on predictability, control and linearity may be insufficient. How TQM approaches are inadequate because they do not address the uncertainties that impact significantly on results in some industries, is also described in [7]. On the other hand, a review of the use of intelligent systems in manufacturing can be found in, e.g. [8]. The review shows variety in the use of these techniques.
Concerning the use of machine learning algorithms for quality management in manufacturing, there are also several approaches. For example, Shigaki and Narazaki [9] demonstrated an approximate summarisation method of process data for acquiring knowledge to improve product quality based on the induction of decision trees, one of machine learning techniques. They also demonstrated a machine learning approach for a sintering process using a neural network [10]. Concerning the ship repair domain there has been no work reported on the use of artificial intelligence for quality management. Thus the use of machine learning algorithms has also not been reported. Instead, approaches based mainly on statistical techniques and ISO 9000 standards can be found (e.g. [11], [12]). On the other hand, some work concerning manufacturing databases in the ship repair domain has been reported (e.g. [13]).
In this study, the approach as suggested within the DQC model is applied. The mechanisms investigated are: (1) systematic recording of data into expertly designed database, (2) standardisation of the data, and (3) transformation of the data into a knowledge base by means of machine learning. The data studied in the research and collected from a real ship repair yard are: (1) parameters defining repair activities that were described within each repair project (attribute values), and (2) related times estimated by the human expert (the target attribute). The data are limited to dock works. The reasons for that are: (1) dock works are technologically self contained subset of repair works, present in almost every ship repair project, (2) dock works often contain activities that influence the overall delivery time the most, such as anti-corrosive and steel works, and (3) since docks appertain to the most valuable and bottleneck resources of any shipyard the duration of these works is always important, and estimated separately. The goal of machine learning from these data was to construct comprehensible delivery time predictors, such as regression or model trees for computer-supported estimate, eliciting the hidden implicit knowledge from the data. Attribute selection and data refinement are done manually, based on the deep understanding of the learning problem and what the attributes actually mean. Given that in the inquiries-answering stage detailed technical data typically are not known, they are not included into this study.
Section snippets
Delivery time estimate in ship repair
The correct estimate of delivery time largely influences the quality and cost of the ship repair service. The delivery time depends generally on factors concerning: (1) the particular works that have to be done within the ship repair project, (2) the features of the shipyard, such as, e.g. physical capacities and capacity loading, facilities, technologies, tools and manpower available, experience and skill of people, (3) delivery time of materials and components, and (4) the situation on the
Machine learning methods and algorithms used
Machine learning is the area of artificial intelligence concerned with the problem of building computer programs that automatically improve with experience [16]. Of the forms of learning, learning concepts from examples is the most common and best understood. Learning from examples is also called ‘inductive learning’. Inductive learning is the most researched kind of learning in artificial intelligence and this research has produced many solid results [17]. In attribute-based supervised
The dock works data model
In order to explore the possibilities of synthesising the time estimating ship repair knowledge, and define the dock works data model, the sample containing 221 examples of ship repair projects was collected. The collected sample corresponds to approximately 70% of inquiries that during one calendar year have usually been received and analysed in that typical medium-sized repair yard. Collected repair projects were described in a shortened non-standardised form on paper, although the shipyard
Learning and analysis of the generated models
Since there is no definite way to choose the most appropriate learning methods in the specific domain [26], in this research several learning tests were carried out. Because of so many possibilities discovered for the reduction of the data dimensionality, tests were performed on different datasets, using different prediction techniques outlined in Section 3. In initial experiments, all attributes were included. After that, attributes estimated as most likely to be irrelevant, redundant or
Discussion and conclusions
Increasingly, differences in a firm's performance are attributed to tacit knowledge (e.g. [27], [28]). According to Simon [29], the reason why experts on a given subject can solve a problem more readily than novices is that the experts have in mind a pattern born on experience, which they can overlay on a particular problem and use to quickly detect a solution. On the other hand, the uncertainty associated with humans increases the need for knowledge formalisation. Rarity of real experts
Acknowledgments
The authors wish to thank the Faculty of Organization and Informatics Varazdin, University of Zagreb, for providing partial funding to support this research. They also owe thanks to the shipyard, and special thanks to Mr. Drago Brzac, the domain expert for his contribution to this research. Thanks also to Mr. Christian Massow from Logimatic, Germany, for his useful suggestions on Section 2, as well as to two anonymous referees who helped us to clarify the positioning of the presented research
Alira Srdoč graduated in Naval Architecture at Technical Faculty, University of Rijeka, Croatia, where she also received her MSc degree. Since now, she has worked in two shipyards, ‘Viktor Lenac’ and ‘3.MAJ’, both in Rijeka, Croatia, in the area of planning, as well as information systems design and implementation. In collaboration with the Artificial Intelligence Laboratory, Jozef Stefan Institut, Ljubljana, Slovenia, she has also developed some industrial knowledge-based systems. Currently
References (31)
- et al.
Integrating quality management practices with knowledge creation processes
Journal of Operation Management
(2004) - et al.
A probabilistic structure life prediction system for container ship repair and inspection
International Journal of Pressure Vessels and Piping
(1992) - et al.
Machine learning approaches to manufacturing
Annals of the CIRP
(1996) - et al.
Using regression trees to identify the habitat preference of the sea cucumber (Holothuria leucospilota) on Rarotonga, Cook Islands
Ecological Modelling
(2003) - et al.
Machine learning approach to machinability analysis
Computers in Industry
(1998) Human capital, shared expertise, and the likelihood of impasse in corporate acquisition
Journal of Management
(2002)Complexity in manufacturing systems
Manufacturing systems (Aachen)
(1995)Knowledge management: re-thinking information management and facing the challenge of managing tacit knowledge
Information Research
(2002)- et al.
A quality management model based on the ‘deep quality concept’
International Journal of Quality and Reliability Management
(2005) - et al.
Process Quality Knowledge Bases
Journal of Quality Management
(2000)
The Paradigms of Quality: Evolution and Revolution in the History of the Discipline. Advances in the Management of Organizational Quality
Managing in uncertainty: lessons from the underground
Quality Progress
Intelligent systems in manufacturing: current developments and future prospects
Integrated Manufacturing Systems
An approximate summarization method of process data for acquiring knowledge to improve product quality
Production Planning & Control
A machine-learning approach for a sintering process using a neural network
Production Planning & Control
Cited by (13)
Deep learning for big data applications in CAD and PLM – Research review, opportunities and case study
2018, Computers in IndustryEffects of meteorological forcing on coastal eutrophication: Modeling with model trees
2012, Estuarine, Coastal and Shelf ScienceCitation Excerpt :MTs have been used to approach pattern, classification, prediction, and hierarchical problems in various research fields. Applications can be found in medical science (Shao et al., 2007; Trujillano et al., 2009), quality management (Srdoc et al., 2007), agriculture (Debeljak et al., 2007; Kocev et al., 2009), water management (Bhattacharya and Solomatine, 2005) and wave forecasting (Etemad-Shahidi and Mahjoobi, 2009; Bonakdar and Etemad-Shahidi, 2011; Jain et al., 2011). Although MTs have been regularly implemented in terrestrial ecology for describing and modeling population dynamics (Stankovski et al., 1998; Demsar et al., 2006; Jurc et al., 2006; Ogris and Jurc, 2010), their applications related to modeling ecological processes in aquatic ecosystems are restricted.
A hybrid approach to design efficient learning classifiers
2009, Computers and Mathematics with ApplicationsCitation Excerpt :For certain kinds of problems, better results may be achieved through this approach as compared to others. Srdoc et al. [32] have shown that the decision tree based approach is suitable for estimating the possible repair time for a ship. Their research indicates also that standardization of problem domain notions and expertly designed databases with possible interfaces to Machine Learning (ML) algorithms (computer algorithms that improve automatically through experience) [33] need to be considered as an integral part of any quality management system.
A fashion mix-and-match expert system for fashion retailers using fuzzy screening approach
2009, Expert Systems with ApplicationsMachine learning for intelligent welding and manufacturing systems: research progress and perspective review
2022, International Journal of Advanced Manufacturing Technology
Alira Srdoč graduated in Naval Architecture at Technical Faculty, University of Rijeka, Croatia, where she also received her MSc degree. Since now, she has worked in two shipyards, ‘Viktor Lenac’ and ‘3.MAJ’, both in Rijeka, Croatia, in the area of planning, as well as information systems design and implementation. In collaboration with the Artificial Intelligence Laboratory, Jozef Stefan Institut, Ljubljana, Slovenia, she has also developed some industrial knowledge-based systems. Currently she is project manager in ‘3.MAJ’, as well as the final year PhD student in Quality Systems at the Department of Control and Manufacturing Systems at the Faculty of Mechanical Engineering, University of Ljubljana, Slovenia. Her research interest is focused on issues related to management, particularly quality and knowledge management, and expert knowledge modelling. She has published several research and professional papers, the most important of which is A quality management model based on the ‘deep quality concept’, published in International Journal of Quality and Reliability Management in 2005.
Ivan Bratko is professor of computer science at Faculty of Computer and Information Science of Ljubljana University, Slovenia. He heads the Artificial Intelligence Laboratory and is also associated with J. Stefan Institute, Slovenia. He has worked as visiting professor at various universities worldwide. His research interests include machine learning, qualitative modelling, computer game playing, and applications of artificial intelligence in biomedicine, ecological modelling and systems control. He has published over 200 research papers, as well as several books, the best known being Prolog Programming for Artificial Intelligence (third ed., Addison-Wesley 2001).
Alojzij Sluga is an associate professor of manufacturing engineering at Faculty of Mechanical Engineering, University of Ljubljana, Slovenia. He heads the Laboratory for Manufacturing Cybernetics and Experimentation. He has worked with different industries in areas of manufacturing technology and computer integrated manufacturing. His current research interest includes enterprise modeling, networked organisations, quality systems and technology. He has published over 100 research papers. He has been scientific responsible for the University of Ljubljana of the EUREKA project TRUST ‘Use of Regional Potentials with Respect to Problem-Solving Processes in Production’, and of the Network of Excellence VRLKCiP ‘Virtual Research Lab for a Knowledge Community in Production’ funded by the European Community. He is a member of SATENA (Slovenian Academic Society for Technology and Natural Science), and associate member of the CIRP (International Academy for Production Engineering).