i4Q Manufacturing Line Reconfiguration Toolkit (i4Q LRT)
i4QLRT is a collection of optimisation micro-services that use simulation to evaluate different possible scenarios and propose changes in the configuration parameters of manufacturing lines to achieve improved quality targets.
Project: i4Q
Publisher: UPV, IKERLAN, EXOS
Technology: Optimisation Algorithms, Machine Learning, Simulation
Industry: Industry
INTRODUCTION
The objective of this solution is to increase productivity and reduce the efforts for line reconfiguration through AI, considering both automated approaches and collaboration with humans. To this end, this solution will provide an optimisation microservice, which will be used to evaluate different possible scenarios and propose changes to the line configuration (including the design, if possible) to achieve improved quality objectives. After the proposed configuration parameters are confirmed, these optimisation microservices will evaluate the new process output characteristics to validate the success of the optimisation and/or adapt its reasoning rules according to the achieved results.
i4Q LRT will use different Python libraries that allow the analysis of data that can be obtained directly from the manufacturing line. Some of the optimisation libraries used are SciPy, network and integer linear programming tools. Thanks to this, using machine learning tools (such as Pytorch or Keras) i4Q LRT will offer different reconfiguration solutions for the production line.
FEATURES/BENEFITS
- Features:
- License
- Full Customisation
- Installation
- Training
- Consultancy and support
- Benefits:
- Identifies strengths and weaknesses and prescribing precise improvement plans, through a full diagnosis of production lines,
- More educated resource allocation and scheduling decisions
- Improve productivity and product quality
- Ensures data quality criteria are met though certification and auditing procedures
TECHNICAL INFORMATION
Code | i4Q LRT | ||
Type | Microservice | ||
Technology Topic | Optimisation Algorithms, Machine Learning, Simulation | ||
Development | YAML, Python | ||
RESOURCES
How it Works | Component | Interact | Identifying technical relationships |
Launch CI/CD Pipeline | Build Distributable File | By using Yaml files for continuous integration, the construction and storage of docker images can be automated | |
Get Description | Show Description | The functionality of obtaining the description of an algorithm is linked to the component of displaying the description. | |
Instantiate Algorithm | Export Algorithm Manage Data Access Configuration | The component of instantiating an algorithm is directly related to exporting it and manage the accessing data. | |
Get algorithm Result | Manage Data Access Configuration | When an algorithm is launched for validation, it requires access to client to obtain data. The Manage Data Access takes care of this configuration. | |
Manage Data Access Configuration | Access Data | The managed data access provides the necessary configuration to the component to be able to access data. The data can be of various formats, be the files, databases, or others. | |
Run Algorithm | View Results | The run algorithm component is linked to the data access component (which requires it to run the algorithm) and the view results component (which displays the result of the algorithm). | |
USE CASE
SERVICES LINKED
PUBLISHERS
UPV - Intending to help to foster the potential of its students and contribute to the transformation of society and its development, the Universitat Politècnica de València focuses its public higher education service on providing the best learning experience to its students, accrediting the competencies and skills acquired for the exercise of professional activities related to art, science and technology, and facilitating lifelong learning with accredited quality training offers.
