Strategizing Inventory Write-Off Process in Sri Lankan Apparel Sector: A Systematic Literature Review

Published: Feb. 21, 2024

The write-off process in the apparel industry is disposal of excess inventory raw materials to optimize inventory. Sri Lanka faces significant challenges optimizing costs, which occur when surplus items are disposed due inaccurate forecasting, changes product order and quantity, misjudgment categories. main reason for these concerns difficulty identifying real-time customers' behaviour, crucial demand forecasting. Our study includes an analysis effects write-offs across diverse industries, alongside exploration determinants influencing processes. A vital challenge encompassed within field rapid variations future orders. This paper provides a detailed studies addressing identification variations, impact optimization on write-offs, methodologies aimed at enhancing this process. primary objective investigation identify existing research gap through comprehensive examination prior literature procedures. evaluates forecasting optimization. We outline factors affecting processes different industries interpret how contributes Through scholarly approach, we provide recommendations strategizing By strategically implementing company can minimize financial losses, reduce streamline operational workflows, positioning itself as market leader while demonstrating commitment sustainability.

Language: Английский

Automobile-Demand Forecasting Based on Trend Extrapolation and Causality Analysis

Electronics, Journal Year: 2024, Volume and Issue: 13(16), P. 3294 - 3294

Published: Aug. 19, 2024

Accurate automobile-demand forecasting can provide effective guidance for automobile-manufacturing enterprises in terms of production planning and supply planning. However, automobile sales volume is affected by historical other external factors, it shows strong non-stationarity, nonlinearity, autocorrelation complex characteristics. It difficult to accurately forecast using traditional models. To solve this problem, a model combining trend extrapolation causality analysis proposed derived from the predictors influence factors. In trend-extrapolation model, series was captured based on Seasonal Autoregressive Integrated Moving Average (SARIMA) Polynomial Regression (PR); then, Empirical Mode Decomposition (EMD), stationarity-test algorithm, an autocorrelation-test algorithm were introduced reconstruct sequence into stationary components with seasonality components, which reduced influences non-stationarity nonlinearity modeling. causality-analysis submodel, 31-dimensional feature data extracted influencing such as date, macroeconomy, promotion activities, Gradient-Boosting Decision Tree (GBDT) used establish mapping between factors future because its excellent ability fit nonlinear relationships. Finally, performance three combination strategies, namely boosting series, stacking parallel weighted-average tested. Comparative experiments groups showed that strategy had best performance, loss reductions 16.81% 4.68% number-one brand, 25.60% 2.79% number-two 46.26% 14.37% number-three brand compared strategies. Other ablation studies comparative six basic models proved effectiveness superiority model.

Language: Английский