Automated Lead Time Estimation for Anomaly Detection using a Machine Learning Algorithm
The increasing complexity of modern business operations demands efficient and accurate lead time estimation to enhance decision-making processes. This study proposes a novel approach to automate lead time estimation using machine learning algorithms. Traditional lead time estimation methods often rely on manual calculations and historical averages, leading to inaccuracies and inefficiencies. In contrast, machine learning algorithms leverage historical data, contextual factors, and patterns to predict lead times dynamically. The automation of lead time estimation not only improves accuracy but also facilitates real-time decision-making. The system continuously learns from new data, adapting its predictions to changing business environments. A user-friendly interface is developed to allow easy input of relevant data and to visualize the lead time prediction. In this project design an automated time estimation is calculated for the usage of two algorithms and to get the accuracy for the maximum amount of iterations to be fitted by the PSO algorithm and then use the K-means clustering for the grouping the classes. From the PSO algorithm get the best features and then apply the Neural Network and Analogy Based Estimation for encrypt the data and then apply model to get the accuracy and time estimation from initialization to the end of the prediction and compare the two model for the accuracy and time comparison and also get the best features.
