Jacek Kozłowski

Comparison of data mining tools for significance analysis of process parameters in applications to process fault diagnosis

This paper presents an evaluation of various methodologies used to determine relative significances of input variables in data-driven models. Significance analysis applied to manufacturing process parameters can be a useful tool in fault diagnosis for various types of manufacturing processes. It can also be applied to building models that are used in process control. The relative significances of input variables can be determined by various data mining methods, including relatively simple statistical procedures as well as more advanced machine learning systems. Several methodologies suitable for carrying out classification tasks which are characteristic of fault diagnosis were evaluated and compared from the viewpoint of their accuracy, robustness of results and applicability. Two types of testing data were used: synthetic data with assumed dependencies and real data obtained from the foundry industry. The simple statistical method based on contingency tables revealed the best overall performance, whereas advanced machine learning models, such as ANNs and SVMs, appeared to be of less value.

A hybrid system with regression trees in steel-making process

The paper presents a hybrid regresseion model with the main emphasis put on the regression tree unit. It discusses input and output variable transformation, determining the final decision of hybrid models and node split optimization of regression trees. Because of the ability to generate logical rules, a regression tree maybe the preferred module if it produces comparable results to other modules, therefore the optimization of node split in regression trees is discussed in more detail. A set of split criteria based on different forms of variance reduction is analyzed and guidelines for the choice of the criterion are discussed, including the trade-off between the accuracy of the tree, its size and balance between minimizing the node variance and keeping a symmetric structure of the tree. The presented approach found practical applications in the metallurgical industry.

Marking and return method for evaluating the effects of stocking larval vendace, Coregonus albula (L.), into Lake Wigry in 2000-2001

Marking and return method for evaluating the effects of stocking larval vendace, Coregonus albula (L.), into Lake Wigry in 2000-2001 In 2000 and 2001, larval vendace, Coregonus albula (L.), were marked and released into Lake Wigry. The larvae were immersed in alizarin red S. Of the 19.2 million vendace larvae released in 2000, 2 million were marked (10.4% of the overall number of fish released), and in 2001 of the 18.8 million fish released, 7 million were marked (47.8%). In subsequent years, otoliths were excised from caught vendace and the number of them with alizarin marks was determined. It was assumed that all unmarked specimens came equally from natural spawning and stocking assuming that survival is equal in both forms of recruitment. It was confirmed that 82.4% of the vendace caught from the 2000 generation originated from stocking, while this figure was 64.2% of all specimens caught from the 2001 generation. Lake Wigry hosts the most abundant vendace population in Poland, and this stock spawns on a massive scale annually. Even so, the study described herein provides evidence of just how important systematic stocking is to the maintenance of the vendace population in this lake. Zastosowanie metody znakowanie - zwroty do oceny efektów zarybień jeziora Wigry larwami sielawy, Coregonus albula (L.) w latach 2000-2001 W roku 2000 i 2001 przeprowadzono znakowanie w alizarynie S larw sielawy, Coregonus albula (L.), przeznaczonych do zarybienia jeziora Wigry. Spośród 19,2 mln larw sielawy w 2000 roku poznakowano 2 mln (10,4% całkowitej ilości wpuszczonych ryb) a w roku następnym z 18,8 mln ryb, poznakowano 7 mln (47,8%). W kolejnych latach od wszystkich odłowionych sielaw pobierano otolity i ustalano ile z nich posiadało znaczek alizarynowy. Założono, że wszystkie osobniki niepoznakowane pochodzą, w równym stopniu, tak z naturalnego tarła, jak i z zarybień, przyjmując podobną przeżywalność ryb z obu źródeł rekrutacji. Stwierdzono, że w odłowach sielaw z pokolenia 2000 ryby z zarybień stanowiły 82,4%, a z pokolenia 2001 64,2% wszystkich odłowionych osobników. Jezioro Wigry jest zbiornikiem o największej liczebnie populacji sielawy w Polsce, w którym co roku odbywa się jej masowe tarło. Jednak wyniki opisywanych badań świadczą o niezbędności systematycznych zarybień dla utrzymania populacji sielawy w jeziorze.

Applications of Data Mining to Diagnosis and Control of Manufacturing Processes

In the majority of manufacturing companies large amounts of data are collected and stored, related to designs, products, equipment, materials, manufacturing processes etc. Utilization of that data for the improvement of product quality and lowering manufacturing costs requires extraction of knowledge from the data, in the form of conclusions, rules, relationships and procedures. Consequently, a rapidly growing interest in DM applications in manufacturing organizations, including the development of complex DM systems, can be observed in recent years (Chen et al. 2004; Chen et al. 2005; Dagli & Lee, 2001; Hur et al., 2006; Malh & Krikler, 2007; Tsang et al., 2007). A comprehensive and insightful characterization of the problems in manufacturing enterprises, as well as the potential benefits from the application of data mining (DM) in this area was presented in (Shahbaz et al., 2006). Examples and general characteristics of problems related to the usage of data mining techniques and systems in a manufacturing environment can be found in several review papers (Harding et al., 2006; Kusiak, 2006; Wang, 2007). Application of DM techniques can bring valuable information, both for designing new processes and for control of currently running ones. Designing the processes and tooling can be assisted by varied computer tools, including simulation software, expert systems based on knowledge acquired from human experts, as well as the knowledge extracted semi automatically by DM methods. The proper choice of the manufacturing process version and its parameters allows to reduce the number of necessary corrections resulting from simulation and/or floor tests. The knowledge obtained by DM methods can significantly contribute to the right decision making and optimum settings of the process parameters. In the design phase two main forms of knowledge may be particularly useful: the decision logic rules in the form: ‘IF (conditions) THEN (decision class)’ and the regression–type relationships. Although the latter have been widely utilized before the emergence of DM methods (e.g. in the form of empirical formulas) and the rules created by the human experts were also in use, the computational intelligence (CI) methods (learning systems) have remarkably enhanced possibilities of the knowledge extraction and its quality. For the manufacturing process control many varied methods are used, ranging from paper Statistical Process Control (SPC) charts to automated closed loop systems. In spite of the

Modeling of Foundry Processes in the Era of Industry 4.0

The paper presents main areas of Industry 4.0 concept with regard to specificity and complexity of foundry processes. Data mining tools are discussed in terms of the possibilities and limitations of their application in Smart Factories. Data acquisition methods are described and the potential areas of restrictions in Internet implementation of things are identified on the example of foundry processes. The methodology of data preparation is also presented, including key tasks and actions to be taken, so that the collected production data are valuable from the point of view of Data Mining tools. As a result, the concept of CPS (Cyber-Physical Systems)/CPPS (Cyber-Physical Production Systems) tool allowing effective implementation of Data Mining tools in complex production processes is presented.

Alizarin mark retention in the otoliths of whitefish (Coregonus lavaretus f. lavaretus L.) from Lake Łebsko, Poland

During the spawning migrations of whitefi sh (Coregonus lavaretus f. lavaretus L.) to Lake Łebsko in Poland during 2009 and 2010, a total of 276 individuals was caught and analysed to determine the retention time of alizarin red S (ARS) marks in their otoliths following marking in 2004 and 2005. ARS marks were visible in the otoliths of 89 specimens, with the oldest fi sh having an age of 5 +. Such mark retention of in excess of 5 years means that bath-marking with ARS is a valuable method for monitoring whitefi sh from Lake Łebsko during a current recovery programme for this endemic population.