Data mining and knowledge discovery (DMKD) focuses on extracting useful information from data. In the chemical process industry, tasks such as process monitoring, fault detection, process control, optimization, etc., can be achieved using DMKD. However, the selection of the appropriate method for each step in the DMKD process, namely data cleaning, sampling, scaling, dimensionality reduction (DR), clustering, clustering analysis and data visualization to obtain meaningful insights is far from trivial. In this contribution, a computational environment (FastMan) is introduced and used to illustrate how method selection affects DMKD in chemical process data. Two case studies, using data from a simulated natural gas liquid plant and real data from an industrial pyrolysis unit, were conducted to demonstrate the applicability of these methodologies in real-life scenarios. Sampling and normalization methods were found to have a great impact on the quality of the DMKD results. Also, a neighbor graphs method for DR, t-distributed stochastic neighbor embedding, outperformed principal component analysis, a matrix factorization method frequently used in the chemical process industry for identifying both local and global changes.

An engineering service project can be highly interactive, collaborative, and distributed. The implementation of such projects needs to generate, utilize, and share large amounts of data and heterogeneous digital objects. The information overload prevents the effective reuse of project data and knowledge, and makes the understanding of project characteristics difficult. Toward solving these issues, this paper emphasized the using of data mining and machine learning techniques to improve the project characteristic understanding process. The work presented in this paper proposed an automatic model and some analytical approaches for learning and predicting the characteristics of engineering service projects. To evaluate the model and demonstrate its functionalities, an industrial data set from the aerospace sector is considered as a the case study. This work shows that the proposed model could enable the project members to gain comprehensive understanding of project characteristics from a multidimensional perspective, and it has the potential to support them in implementing evidence-based design and decision making.

An important task of knowledge discovery deals with discovering association rules. This very general model has been widely studied and efficient algorithms have been proposed. But most of the time, only frequent rules are seeked. Here we propose to consider this problem as a multi-objective combinatorial optimization problem in order to be able to also find non frequent but interesting rules. As the search space may be very large, a discussion about different approaches is proposed and a hybrid approach that combines a metaheuristic and an exact operator is presented.

Industrial robots in poorly structured environments have to interact compliantly with this environment for successful operations. In this paper, we present a behaviour-based approach to learn peg-in-hole operations from scratch. The robot learns autonomously the initial mapping between contact states to motion commands employing fuzzy rules and creating an Acquired-Primitive Knowledge Base (ACQ-PKB), which is later used and refined on-line by a Fuzzy ARTMAP neural network-based controller. The effectiveness of the approach is tested comparing the compliant motion behaviour using the ACQ-PKB and a priori Given-Primitive Knowledge Base (GVN-PKB). Results using a KUKA KR15 industrial robot validate the approach.