The purpose of this thesis is twofold. On the one hand, it is proposed a simplified optimisation model aimed at reducing the stock time of parts in the warehouse. On the other hand, it is analysed a case found at Wärtsilä Italia in order to highlight the problems inherent in batching and how these, together with the effects generated by Selective Assembly, that may be verified when trying to reduce the variability on the lead time of the parts in stock, considering an input policy based on make to stock, an output policy based on make to order. In order to develop the simplified model, the principles of Selective Assembly were used and the batching of components in stock was not considered, in order to highlight the improvements that a correct sizing of the number and range of classes can generate on the stocking time of components in stock over a given period of time, making a comparison with the configuration currently used by the company. The paper also compares the configuration proposed by the model with configurations based on the principle of equiprobable classes, using the same input data, in order to ensure further validity for the model. Initially, an introduction is made to Wärtsilä Italia, the company with which the paper was produced. This is followed by a detailed description of the company problem that was used as a starting point for this project. In the central phase of the work, a detailed analysis of the state of the art of Selective Assembly is first carried out, in which the main classification models and the various optimisation approaches are reported. This is followed by a description of the simplified model, defining the assumptions, variables and constraints imposed by the assembly and required to linearise the problem, comparing them with those imposed by the system should batching be considered. Following this description, a digression between the objective function used by the simplified model and the variable lead time is developed, underlining the positive and negative aspects of the two functions. At the conclusion of this central part, the various steps leading to the validation of the model are reported. The paper concludes with a comparison between the chosen configuration and the one currently implemented by the company. The results obtained show a reduction in component stocking times, comparable to those obtained with the equiprobability classes. Therefore, these results confirm the improvement made by the model presented in the thesis and provide inspiration for further developments and applications.
Lo scopo di questa tesi è duplice. Da una parte, si vuole implementare un modello semplificato di ottimizzazione che miri a ridurre il tempo di giacenza dei pezzi a magazzino. Dall’altra, si vuole analizzare un caso riscontrato presso Wärtsilä Italia in modo da evidenziare quali sono i problemi inerenti al batching e come questi, insieme agli effetti generati dal Selective Assembly, che si possono verificare quando si vuole ridurre la variabilità sul lead time dei pezzi a magazzino, considerando una policy di ingresso basata sul make to stock, una policy di uscita basata sul make to order. Per sviluppare il modello semplificato sono stati utilizzati i principi del Selective Assembly e non è stato considerato il batching dei componenti a magazzino, per poter evidenziare i miglioramenti che un corretto dimensionamento del numero e del range delle classi può generare sul tempo di giacenza dei componenti a magazzino in un periodo di tempo determinato, facendo un confronto con la configurazione attualmente usata dall’azienda. L’elaborato, inoltre, confronta la configurazione proposta dal modello semplificato con le configurazioni basate sul principio di classi equiprobabili, utilizzando gli stessi dati di input, in modo da garantire un ulteriore validità al modello. Inizialmente si presenta Wärtsilä Italia, azienda con la quale si è collaborato per la realizzazione dell’elaborato. Segue una descrizione dettagliata del problema aziendale che è stato utilizzato come punto di partenza di questo progetto. Nella fase centrale del lavoro viene prima svolta un’analisi dettagliata dello stato dell’arte del Selective Assembly, in cui vengono riportati i principali modelli di classificazione e i vari approcci di ottimizzazione. A seguire viene descritto il modello semplificato, definendo le ipotesi, le variabili e i vincoli imposti dall’assemblaggio e necessari per linearizzare il problema, confrontandoli con quelli imposti da sistema qualora si volesse considerare il batching. A seguito di questa descrizione viene sviluppata una digressione tra la funzione obbiettivo utilizza dal modello semplificato e la variabilità del lead time, sottolineando gli aspetti positivi e negativi delle due funzioni. In conclusione di questa parte centrale, sono riportate le varie fasi che hanno portato alla validazione del modello. L’elaborato si conclude con il confronto tra la configurazione scelta e quella attualmente implementata dall’azienda. I risultati ottenuti mostrano una riduzione dei tempi di stoccaggio dei componenti, paragonabili a quelli ottenuti con le classi equiprobabili. Pertanto, tali esiti, consento di confermare il miglioramento apportato dal modello presentato nella tesi e danno spunto per ulteriori sviluppi e applicazioni.
Optimization of classes configuration to reduce lead time in stock
Visconti, Davide Pio
2022/2023
Abstract
The purpose of this thesis is twofold. On the one hand, it is proposed a simplified optimisation model aimed at reducing the stock time of parts in the warehouse. On the other hand, it is analysed a case found at Wärtsilä Italia in order to highlight the problems inherent in batching and how these, together with the effects generated by Selective Assembly, that may be verified when trying to reduce the variability on the lead time of the parts in stock, considering an input policy based on make to stock, an output policy based on make to order. In order to develop the simplified model, the principles of Selective Assembly were used and the batching of components in stock was not considered, in order to highlight the improvements that a correct sizing of the number and range of classes can generate on the stocking time of components in stock over a given period of time, making a comparison with the configuration currently used by the company. The paper also compares the configuration proposed by the model with configurations based on the principle of equiprobable classes, using the same input data, in order to ensure further validity for the model. Initially, an introduction is made to Wärtsilä Italia, the company with which the paper was produced. This is followed by a detailed description of the company problem that was used as a starting point for this project. In the central phase of the work, a detailed analysis of the state of the art of Selective Assembly is first carried out, in which the main classification models and the various optimisation approaches are reported. This is followed by a description of the simplified model, defining the assumptions, variables and constraints imposed by the assembly and required to linearise the problem, comparing them with those imposed by the system should batching be considered. Following this description, a digression between the objective function used by the simplified model and the variable lead time is developed, underlining the positive and negative aspects of the two functions. At the conclusion of this central part, the various steps leading to the validation of the model are reported. The paper concludes with a comparison between the chosen configuration and the one currently implemented by the company. The results obtained show a reduction in component stocking times, comparable to those obtained with the equiprobability classes. Therefore, these results confirm the improvement made by the model presented in the thesis and provide inspiration for further developments and applications.File | Dimensione | Formato | |
---|---|---|---|
Thesis_Visconti_Davide_Pio.pdf
accessibile in internet solo dagli utenti autorizzati
Dimensione
1.27 MB
Formato
Adobe PDF
|
1.27 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/201745