![](/style/images/tocco.png)
Maria Teresa Manente
Development and Scaling of a smart control system to multiple products and multiple manufacture sites.
Rel. Roberto Fontana, Gueorgui Mihaylov. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Gestionale (Engineering And Management), 2024
|
PDF (Tesi_di_laurea)
- Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives. Download (3MB) | Preview |
Abstract: |
This thesis outlines the key aspects of developing and scaling a smart control system to integrate multiple products and manufacture sites based on an analysis performed in collaboration with the Data Science Department of Haleon UK. Nowadays consumers are looking for a broader variety of oral healthcare and in particular toothpastes. This trend towards products increasingly aligned with their needs contributes to an ever-increasing demand curve. The manufacturing processes of toothpastes are highly complex, involving key steps such as ingredient addition, mixing, control of pressure, temperature, and speeds. To proactively monitor these manufacturing processes, Haleon, a multinational consumer health care company, developed a nearly real-time smart control system that guarantees an optimal output not only in terms of product specifications but also in terms of process time, costs and resources. In today’s business world, digital solutions play a crucial role in improving the operational efficiency and facilitating analysis and more sophisticated data management. This thesis project is representative of a standard real-world industrial problem of rapidly building and scaling a digital solution while accelerating adoption and delivery value. Effectively managing a broad and diversified production requires the control system to be easily and quickly scalable to integrate multiple products and manufacture sites. The thesis aims to identify the most relevant considerations faced during the development and scaling process of a control system from both technical and management standpoint. From a technical point of view, the thesis exposes not only the core of the data science model, which is the Fast Dynamic Time Warping, a machine learning algorithm that allows the proactive detection of steps of different processes and production sites, but it also provides an overview of the digital architecture that must be robust and well-designed to ensure that the different bottlenecks of the manufacture sites and requirements of all processes involved are taken into account. On the other hand, from a management perspective, given the high specificity and stringent quality standards of the manufacturing processes involved, the solution is developed in-house. The main management challenge lies in developing and simultaneously implementing an advanced digital solution with the aim of achieving accelerated value generation. This challenge is facilitated by taking into account the needs of all stakeholders and by the accurate and informed decisions made throughout the project lifecycle (from the Proof of Concept up to the scaling phase) with a focus on the minimum viable produce (MVP), a simplified version of the final solution already containing all the key functionalities which reduces the release time and facilitates the scaling process. |
---|---|
Relatori: | Roberto Fontana, Gueorgui Mihaylov |
Anno accademico: | 2023/24 |
Tipo di pubblicazione: | Elettronica |
Numero di pagine: | 106 |
Soggetti: | |
Corso di laurea: | Corso di laurea magistrale in Ingegneria Gestionale (Engineering And Management) |
Classe di laurea: | Nuovo ordinamento > Laurea magistrale > LM-31 - INGEGNERIA GESTIONALE |
Aziende collaboratrici: | Haleon UK Trading Limited |
URI: | http://webthesis.biblio.polito.it/id/eprint/32062 |
![]() |
Modifica (riservato agli operatori) |