ANALISA AKAR MASALAH RADIAL RUN OUT BAN MENGGUNAKAN DECISION TREE

Problem solving in the multistage production process is a challenge for the industry. The use of modern techniques such as machine learning in solving quality problems continues to be developed. One of the machine learning is decision tree. The tire industry entered the era of industrial revolution 4.0 with the use of information technology. Utilizing data using machine learning in finding the root cause of the problem can support the tire industry in industrial competition. This study aims to explore the process data in the tire industry to solve one of the tire quality problems, namely radial run-out tires. The technique of finding the root of the problem in this research is done using Classification and Regression Tree (CART) technique. Input variables involve 60 factors in the production process. From the research, it was found that the factors that influence the radial run out value are the lot of the Tread, Bead and Sidewall components. The factors causing the high radial run-out of the tires are the variations in the lot of the tire components Tread and Bead. The decision tree model that was formed has a precision level of 74.7% in detecting high radial run-out events. The effects of improvement on the lot tread and bead components resulting from the decision tree can reduce the defect of radial run out rate by 99.9%. Keywords: Decision tree; Root cause analysis; Radial run-out Tire; Data mining AbstrakPemecahan masalah pada proses produksi multistage merupakan tantangan untuk indusri. Pemanfaatan teknik modern seperti machine learning dalam pemecahan masalah kualitas terus dikembangkan. Salah satu machine learning adalah decision tree. Industri ban memasuki era industri revolusi 4.0 dengan adanya pemakaian teknologi informasi seperti barcode atau radio frequency identification. Pemanfaataan data dengan menggunakan machine learning dalam pencarian akar masalah bisa mendukung industri ban dalam kompetisi industri. Penelitian ini bertujuan untuk mengekplorasi data proses pada industri ban untuk memecahkan permasalahan kualitas ban yaitu radial run-out ban. Teknik pencarian akar masalah dilakukan menggunakan Clasification and Regression Tree (CART). Variabel input melibatkan 60 faktor dalam proses produksi. Dari penelitian didapatkan faktor yang mempengaruhi nilai radial run out adalah lot komponen Tread, Bead dan Sidewall. Untuk faktor penyebab tingginya radial run-out ban adalah variasi lot komponen Tread dan Bead. Model decision tree yang terbentuk memiliki tingkat presisi 74,7% dalam mendeteksi kejadian radial run-out berkategori tinggi. Efek perbaikan pada komponen lot Tread dan Bead yang dihasilkan dari decision tree dapat menurunkan tingkat defect radial run- out ban sebesar 99,9%.

Analysis of the Possibilities of Tire-Defect Inspection Based on Unsupervised Learning and Deep Learning

At present, inspection systems process visual data captured by cameras, with deep learning approaches applied to detect defects. Defect detection results usually have an accuracy higher than 94%. Real-life applications, however, are not very common. In this paper, we describe the development of a tire inspection system for the tire industry. We provide methods for processing tire sidewall data obtained from a camera and a laser sensor. The captured data comprise visual and geometric data characterizing the tire surface, providing a real representation of the captured tire sidewall. We use an unfolding process, that is, a polar transform, to further process the camera-obtained data. The principles and automation of the designed polar transform, based on polynomial regression (i.e., supervised learning), are presented. Based on the data from the laser sensor, the detection of abnormalities is performed using an unsupervised clustering method, followed by the classification of defects using the VGG-16 neural network. The inspection system aims to detect trained and untrained abnormalities, namely defects, as opposed to using only supervised learning methods.

End-of-life tire management: a critical review

Environmental and social awareness are the key elements of the sustainable tire industry. End-of-life tire (ELT) waste flow is an important environmental problem worldwide since it produces severe air, water, and soil pollution issues. Significant advancements have been made in ELT management in the last few years. As a result, ELTs should not only be regarded as waste but also as a source of environmentally friendly materials. Besides, sound ELT management has vital importance for circular economy and sustainable development. Over the last decade, ELT management has attracted many researchers and practitioners. Unfortunately, a comprehensive review of the ELT management area is still missing. This study presents the first critical review of the whole ELT management area. It aims to present an extensive content analysis overview of state-of-the-art research, provide its critical analysis, highlight major gaps, and propose the most significant research directions. A total of 151 peer-reviewed studies published in the journals between 2010–2020 are collected, analyzed, categorized, and critically reviewed. This review study redounds comprehensive insights, a valuable source of references, and major opportunities for researchers and practitioners interested in not only ELT material flow but also the whole waste management area. Graphical abstract

Unsupervised Learning with Generative Adversarial Network for Automatic Tire Defect Detection from X-ray Images

Automatic defect detection of tire has become an essential issue in the tire industry. However, it is challenging to inspect the inner structure of tire by surface detection. Therefore, an X-ray image sensor is used for tire defect inspection. At present, detection of defective tires is inefficient because tire factories commonly conduct detection by manually checking X-ray images. With the development of deep learning, supervised learning has been introduced to replace human resources. However, in actual industrial scenes, defective samples are rare in comparison to defect-free samples. The quantity of defective samples is insufficient for supervised models to extract features and identify nonconforming products from qualified ones. To address these problems, we propose an unsupervised approach, using no labeled defect samples for training. Moreover, we introduce an augmented reconstruction method and a self-supervised training strategy. The approach is based on the idea of reconstruction. In the training phase, only defect-free samples are used for training the model and updating memory items in the memory module, so the reproduced images in the test phase are bound to resemble defect-free images. The reconstruction residual is utilized to detect defects. The introduction of self-supervised training strategy further strengthens the reconstruction residual to improve detection performance. The proposed method is experimentally proved to be effective. The Area Under Curve (AUC) on a tire X-ray dataset reaches 0.873, so the proposed method is promising for application.

Circular Green Technology & Material for the Tire Industry

The whole Tire Industry around the Globe is set on an important mission to create a greener environment wherein the used tires, scrap worn out tires & shop floor rejected tires are used back in to the system of new Tire manufacturing thereby create a Circular Economy in the Tire Industry via non-chemical Devulcanization process. The Tyromer TDP (Tyre Derived Polymer) production process uses an industrial proven extrusion technology in a patented Twin Screw Extruder and it is reliable. The process is energy efficient as it is continuous. That also gives fundamentally more consistent product quality compared to batch processes. In this extrusion process, what goes in must come out and hence the TDP production process creates no waste. The only catalyst used in the process is Super Critical Carbon dioxide. No chemical solvents or devulcanization chemicals are used and the process is Energy efficient (400 kWh/MT), Very Fast (2 minutes from crumb powder to TDP) and having High conversion rate (99+% crumb powder to TDP).

Evaluation of natural and epoxidized vegetable oil in elastomeric compositions for tread rubber

In the tire industry, the incorporation of natural origin oils in the development of elastomeric formulations has been one of the alternatives to reduce the use of petroleum derivatives, with a high content of toxic compounds. In this work, soybean vegetable oil was investigated as a lubricant and co-activator in sulfur-vulcanized natural rubber compounds. The soybean oil was used in its natural state and chemically modified by the epoxy ring’s introduction in its structure. In an internal mixer a standard formulation of natural rubber, five formulations replacing a conventional aromatic oil and stearic acid by vegetable oil, and a formulation without an activation system were prepared. The natural and epoxidized soybean oil was characterized chemically, and the elastomeric compositions were evaluated by mechanical and rheological analysis. The mechanical properties showed satisfactory results when vegetable soybean oil was used as a lubricant and could be a substitute for conventional aromatic oils, thus guaranteeing reduction of aromatic polycyclic content in the formulations. The crosslink degree and the rheological characteristics of the samples prepared with vegetable soybean oil were similar to the natural rubber standard sample. The formulations without the zinc oxide and stearic acid evidenced the need for activators in the vulcanization reaction, as they presented properties below standard. We verified that the epoxidized soybean oil, even when promoting better dispersion of the fillers, interfered in the crosslink formation, and consequently there was a decrease in the mechanical properties of these formulations. Finally, we indicated vegetable soybean oil as a substitute for aromatic oil and stearic acid, in the elastomeric compositions used to manufacture treads.

EFFECT OF EMULSION SBR PREPARED BY ASYMMETRIC REVERSIBLE ADDITION-FRAGMENTATION TRANSFER AGENT ON PROPERTIES OF SILICA-FILLED COMPOUNDS

ABSTRACT The development of ultra-high-performance tires that satisfy fuel efficiency, traction, handling performance, and abrasion resistance has gained significant importance in the tire industry. Solution SBR has been used as a raw material, owing to its useful characteristics (e.g., narrow dispersity controllable microstructure and chain-end functionalization). In a recent improvement, emulsion SBR (ESBR), a high-molecular-weight compound with narrow dispersity, has been reported for application in the tire tread compounds. In particular, S,S-dibenzyl trithiocarbonate (DBTC) reversible addition-fragmentation transfer (RAFT) ESBR has exhibited excellent abrasion resistance and fuel efficiency in unfilled and carbon black–filled vulcanizates. However, owing to the symmetrical structure of DBTC RAFT ESBR, the polymer chain was shortened by the reaction of a silane coupling agent with trithiocarbonate, leading to poor abrasion resistance and fuel efficiency in the case of silica-filled vulcanizates. In this study, benzyl (4-methoxyphenyl) trithiocarbonate (BMPTC), an asymmetric RAFT agent that promotes unilateral polymer growth, was synthesized and used in the polymerization of BMPTC RAFT ESBR. Chain cleavage was not observed. Upon application to silica-filled vulcanizates, BMPTC RAFT ESBR exhibited improved abrasion resistance (by 9%), improved fuel efficiency (by 20%), and improved wet traction performance (by 10%) compared with the DBTC RAFT ESBR.

Scrap tire management in Ontario

This project paper addresses the major problem of scrap tire management in Ontario. It examines the environmental and health impacts associated with current disposal practices. To address the absence of a management program and lack of regulation for tire disposal in Ontario, tire management programs implemented in other provinces in Canada are evaluated and compared with Ontario, in order to explore the causes of its failure to develop and implement a comprehensive scrap tire management program. The divergence from Ontario on adopting a similar kind of tire managenet program is because of its strong market driven tire industry and governent's reluctance to implement another tax. This paper encourages a tire management system for Ontario as well, which is very similar to those already implemented in other provinces throughout Canada.