Implementing statistical methods to solve DDMRP main issues
This master thesis analyses the supply chain methodology called DDMRP and its known issues. However, DDMRP presents two main matters for the stock planners of the companies. The first one is the computation of the ADU and the second one is the sizing of the Red Zone of the Stock Buffers. The goal of...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Fecha de publicación: | 2022 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/368989 |
| Acceso en línea: | https://hdl.handle.net/2117/368989 |
| Access Level: | acceso abierto |
| Palabra clave: | Mathematical statistics -- Applications DDMRP ARIMA Prophet Holt-Winters Montecarlo Supply Chain Stock Estadística matemàtica -- Aplicacions Classificació AMS::62 Statistics::62P Applications Àrees temàtiques de la UPC::Matemàtiques i estadística::Estadística matemàtica |
| Sumario: | This master thesis analyses the supply chain methodology called DDMRP and its known issues. However, DDMRP presents two main matters for the stock planners of the companies. The first one is the computation of the ADU and the second one is the sizing of the Red Zone of the Stock Buffers. The goal of the study is to find alternatives to these two issues. To compute the ADU more accurately, three time series methodologies are proposed: ARIMA, Holt-Winters and Prophet. On the other hand, the Red Zone size is computed minimizing a formulated cost function. Using a Montecarlo simulation, the number of stock units that minimizes the cost function can be found. With this procedure the planner would not need to establish any parameter, only known values of the product. |
|---|