دانلود رایگان مقاله انگلیسی سیستم پشتیبانی تصمیم برای مدیریت زنجیره تامین مواد غذایی تازه - تیلور و فرانسیس 2018

The paper proposes a decision support system (DSS) for the supply chain of packaged fresh and highly perishable products. The DSS combines a unique tool for sales forecasting with order planning which includes an individual model selection system equipped with ARIMA, ARIMAX and transfer function forecasting model families, the latter two accounting for the impact of prices. Forecasting model parameters are chosen via two alternative tuning algorithms: a two-step statistical analysis, and a sequential parameter optimisation framework for automatic parameter tuning. The DSS selects the model to apply according to user-defined performance criteria. Then, it considers sales forecasting as a proxy of expected demand and uses it as input for a multi-objective optimisation algorithm that defines a set of non-dominated order proposals with respect to outdating, shortage, freshness of products and residual stock. A set of real data and a benchmark – based on the methods already in use – are employed to evaluate the performance of the proposed DSS. The analysis of different configurations shows that the DSS is suitable for the problem under investigation; in particular, the DSS ensures acceptable forecasting errors and proper computational effort, providing order plans with associated satisfactory performances.

4. Conclusions

In this paper, we propose a modular and reliable DSS for sales forecasting and order planning in the supply chain management of packaged fresh food products. The proposed DSS combines, in a unique, flexible and easy-to-use software tool, a forecasting module to derive sales forecasts from historical data and exogenous variables, supported by a model selection and tuning module for the automatic choice and configuration of the forecasting method, and a multi-objective optimisation module equipped with an order plan selection module to derive the best order proposal based on a set of KPIs accounting for cost and quality of service. Three different forecasting model families were considered and tested on a set of sample products in a real supply network.

Our results clearly show the benefits of deriving an optimal order proposal based on sales forecasting, explicitly accounting for demand variability and the possible impact of exogenous variables. The proposed analyses highlight the capability of the DSS to absorb relevant differences in terms of forecasting behaviour, thus limiting their impact on the order planning phase. Another advantage offered by this DSS relies on its flexibility, as it is designed to be easy-to-use and to automatically run alternative approaches in terms of forecasting and model tuning techniques, depending on the characteristics of the data-set. In this respect, we notice that the results point out that there is no dominant forecasting model and there is no convenience to use a single model for all the cases (i.e. pairs item/store), and also the performance of a model selected for a specific case might deteriorate over time. Hence, instead of a one-size-fits-all approach, an individual selection including the identification of the best method for each series is considered, though it is more computationally intensive. More specifically, configurations using SPO tend to provide more accurate forecasts, although the computational time, when tuning is required, is higher than the configurations adopting the grid search.