| dc.description.abstract | Demand forecasting and production planning are the two key factors which contribute to
the success of a supply chain business. Nowadays, with the explosive growth in demand for
fashion, especially footwear. Therefore, shortening material preparation, operation and
delivery time is one of the considerable challenges for companies to attract customers. In
this thesis, two main objectives are proposed to complete. First of all, a machine learning
model using Extremely Gradient Boosting algorithm is built to predict the customer demand
1 month in advance. Then, the forecast result will be the output for the operation to produce
the product. At that phase, a job-shop scheduling model is generated to optimize the total
completion time in operation. In this study, the number of workstations, machines, and
labors are considered to reduce the tardiness of both workers and machines. The model is
applied in a footwear company. However, it can be useful for any industry which uses jobshop scheduling in production. | en_US |
