Muvu Technologies

Abstract

The advent of large Machine Learning models and the steep increase in the demand for AI solutions occurs at the same point in time in which policies are being enacted to implement more sustainable processes in virtually every sector. This means there is a need for more, better and larger models, which require significant computational resources, while at the same time a call for a decrease in the energy spent in the processes associated to MLOps. In this paper we propose a reduced set of meta-features that can be used to characterize sets of data and their relationship with model performance. We start from a large set of 66 features, and reduce it to only 10 while maintaining the strength of this relationship. This ensures a process of meta-feature extraction and prediction of model performance that is in line with the desiderata of Frugal AI, allowing to develop more efficient ML processes.

Journal: Advances in Artificial Intelligence in Manufacturing II

Read the full article: link.

Keywords: Frugal AI, MLOps, Meta-Learning, Manufacturing

Abstract

The rapid integration of Machine Learning (ML) in organizational practices has driven demand for substantial computational resources, incurring both high economic costs and environmental impact, particularly from energy consumption. This challenge is amplified in dynamic data environments, where ML models must be frequently retrained to adapt to evolving data patterns. To address this, more sustainable Machine Learning Operations (MLOps) pipelines are needed for reducing environmental impacts while maintaining model accuracy. In this paper, we propose a model reuse approach based on data similarity metrics, which allows organizations to leverage previously trained models where applicable. We introduce a tailored set of meta-features to characterize data windows, enabling efficient similarity assessment between historical and new data. The effectiveness of the proposed method is validated across multiple ML tasks using the cosine and Bray–Curtis distance functions, which evaluate both model reuse rates and the performance of reused models relative to newly trained alternatives. The results indicate that the proposed approach can reduce the frequency of model retraining by up to 70% to 90% while maintaining or even improving predictive performance, contributing to more resource-efficient and sustainable MLOps practices.

Journal: Big Data and Cognitive Computing

Read the full article: link.

Keywords: MLOps, Frugal AI, Machine Learning and Meta-Learning

Abstract

The technological evolution carried out in recent years has enabled significant developments in various areas of Artificial Intelligence (AI), such as Generative AI. Large Language Models (LLMs) are becoming increasingly complex, allowing for better results and enhancing their real-world applicability. However, these models still face issues such as hallucination or outdated information. This last one occurs due to the temporal gap between the training process and the model’s use. A Retrieval-Augmented Generation (RAG) architecture can address these issues since the information source used is not involved in the training phase, which also facilitates the reuse of models for different applications. One of the challenges of RAG is its applicability when the data source is a relational database, becoming even more challenging as the database size and complexity increase. This article proposes a potential architecture and approach for solving this problem and implementing a RAG architecture using a relational database as the data source.

Conference: 2024 8th International Symposium on Innovative Approaches in Smart Technologies (ISAS)

Read the full article: link.

Keywords: RAG, LLM, Generative AI

Abstract

Many of today’s domains of application of Machine Learning (ML) are dynamic in the sense that data and their patterns change over time. This has a significant impact in the ML lifecycle and operations, requiring frequent model (re-)training, or other strategies to deal with outdated models and data. This need for dynamic and responsive solutions also has an impact on the use of computational resources and, consequently, on sustainability indicators. This paper proposes an approach in line with the concept of Frugal AI, whose main aim is to minimize the resources and time spent on training models by re-using models from a pool of past models, when appropriate. Specifically, we present and validate a methodology for similarity-based model selection in data streaming environments with concept drift. Rather than training a new model for each new block of data, this methodology considers a pool with only a subset of the models and, for each new block of data, will select the best model from the pool. The best model is determined based on the distance between its training data and the current block of data. Distance is calculated based on a set of meta-features that characterizes the data, and on the Bray-Curtis distance. We show that it is possible to reuse previous models using this methodology, leading to potentially significant saving of resources and time, while maintaining predictive quality.

Conference: ISAmI – 15th International Symposium on Ambient Intelligence

Keywords: Machine Learning, Meta-Feature Extraction, Concept Drift, Model Selection, Frugal AI

Abstract

Globally, sustainability concerns have been driving several green strategies towards new business models for industry circularity. While industrial greenality is broadly discussed in literature, still missing a clear roadmap tailored for small and medium enterprises (SMEs) to translate circularity practices into tangible, value-driven datasets. This research work addresses this existing gap by proposing a circularity information management roadmap designed for SMEs. The framework aims to guide small enterprises throughout their journey on the implementation of circular business practices. The discussed framework intends to be used as a governance model to assess and increase an organization’s circular level. By providing a clear and common view on circular practices and its underlying artefacts, SMEs might design their circular projects more efficiently based on a clear understanding on the needs and means to increase their circularity business performance.

Conference: CISTI’2024 – 19th Iberian Conference on Information Systems and Technologies

Keywords: Circular Economy, Information Management, Circularity Governance Model, Circularity Roadmap

Abstract

The fourth industrial revolution has ushered in a transformative era for the industrial sector, marked by the integration of advanced technologies that increase production efficiency and quality. Among these innovations is Smart Scheduling, which aims to optimize production processes while minimizing costs and meeting manufacturing requirements. This article explores the development and application of a Hybrid Genetic Algorithm, optimized by Tabu Search, to meet the complex challenge of multi-objective industrial production planning. The proposed framework introduces a Genetic Algorithm designed for intelligent, multi-objective industrial production planning, demonstrating its effectiveness in generating high-quality planning. It significantly reduces production times while maintaining quality and business needs, as validated in a real-world case study. The algorithm improves on traditional planning methods by dynamically adapting to manufacturing priorities. It also describes potential future adaptations to a wider range of industrial contexts and sectors. This research highlights the fundamental role of genetic algorithms in advancing smart manufacturing, offering a scalable and adaptable solution to the challenges of Industry 4.0.

Conference: ECAI 2024 16th Edition – INTERNATIONAL CONFERENCE on Electronics, Computers and Artificial Intelligence

Read the full article: link.

Keywords: Industry 4.0, Genetic Algorithm, Tabu Search, Multi-Objective Optimization, Smart Scheduling

Abstract

With the spread of the Industry 4.0 concept, implementing Artificial Intelligence approaches on the shop floor that allow companies to increase their competitiveness in the market is starting to be prioritized. Due to the complexity of the processes used in the industry, the inclusion of a real-time Quality Prediction methodology avoids a considerable number of costs to companies. This paper exposes the whole process of introducing Artificial Intelligence in plastic injection molding processes in a company in Portugal. All the implementations and methodologies used are presented, from data collection to real-time classification, such as Data Augmentation and Human-in-the-Loop labeling, among others. This approach also allows predicting and alerting with regard to process quality loss. This leads to a reduction in the production of non-compliant parts, which increases productivity and reduces costs and environmental footprint. In order to understand the applicability of this system, it was tested in different injection molding processes (traditional and stretch and blow) and with different materials and products. The results of this document show that, with the approach developed and presented, it was possible to achieve an increase in Overall Equipment Effectiveness (OEE) of up to 12%, a reduction in the process downtime of up to 9% and a significant reduction in the number of non-conforming parts produced. This improvement in key performance indicators proves the potential of this solution.

Read the full article: link.

Keywords: Artificial Intelligence; Predictive quality; Injection molding; Data Augmentation; Human-in-the-Loop labeling; OEE