Authors :
Abirami R; Deepika Sanga; Sowmiya R; Mohd Amer Hussain; Bharani Kumar Depuru
Volume/Issue :
Volume 9 - 2024, Issue 3 - March
Google Scholar :
https://tinyurl.com/tv889f27
Scribd :
https://tinyurl.com/mtm2wp3x
DOI :
https://doi.org/10.38124/ijisrt/IJISRT24MAR1939
Abstract :
This study addresses the major challenges of
forecasting automotive kit items(parts of vehicles) by
enhancing the delivery of the products and managing the
inventory. The kit items vary as per customers and it is
unique on its own, where the uniqueness determines the
vehicle parts. Customers are the major role players who
provide the business hence, this study highlights various
factors contributing to the customer’s choice of kit items
with features consisting of vehicle name, original
equipment manufacturer (OEM), Item Description
(collection of vehicle parts) type of product (brand of
vehicle) and monthly allotment of each kit item as per
customer starting from 2021 April to 2024 January.
We conducted an extensive analysis to assess a range
of time series analysis techniques for predicting kit
demand within the automotive industry, the methods we
investigated encompassed Autoregressive (AR),
Autoregressive Moving Average (ARMA)
,Autoregressive Integrated Moving Average (ARIMA),
Seasonal Autoregressive Integrated Moving Average
(SARIMA), Simple Exponential Smoothing (SES), Holt's
Linear Trend Method - Double Exponential Smoothing,
Triple Exponential Smoothing - Holt Winters, Long
Short-Term Memory (LSTM) and advanced forecasting
models such as prophet in evaluating the accuracy of these
models, we employed key metrics such as Root Mean
Squared Error (RMSE) and Mean Absolute Percentage
Error (MAPE), this study aims to drive significant
progress in the automotive industry by optimising
inventory management reducing storage costs and
improving delivery efficiency to ensure smooth business
operations moreover the integration of visually engaging
dashboards for real-time analysis of projected values
plays a pivotal role in identifying crucial monthly demand
trends this integration not only enhances operational
efficiency but also fosters enriched customer engagement
thereby facilitating sustained advancement within the
automotive sector.
Keywords :
Time Series Analysis, Demand Forecasting, Inventory Management, Deep Learning, Prophet, Supply Chain.
This study addresses the major challenges of
forecasting automotive kit items(parts of vehicles) by
enhancing the delivery of the products and managing the
inventory. The kit items vary as per customers and it is
unique on its own, where the uniqueness determines the
vehicle parts. Customers are the major role players who
provide the business hence, this study highlights various
factors contributing to the customer’s choice of kit items
with features consisting of vehicle name, original
equipment manufacturer (OEM), Item Description
(collection of vehicle parts) type of product (brand of
vehicle) and monthly allotment of each kit item as per
customer starting from 2021 April to 2024 January.
We conducted an extensive analysis to assess a range
of time series analysis techniques for predicting kit
demand within the automotive industry, the methods we
investigated encompassed Autoregressive (AR),
Autoregressive Moving Average (ARMA)
,Autoregressive Integrated Moving Average (ARIMA),
Seasonal Autoregressive Integrated Moving Average
(SARIMA), Simple Exponential Smoothing (SES), Holt's
Linear Trend Method - Double Exponential Smoothing,
Triple Exponential Smoothing - Holt Winters, Long
Short-Term Memory (LSTM) and advanced forecasting
models such as prophet in evaluating the accuracy of these
models, we employed key metrics such as Root Mean
Squared Error (RMSE) and Mean Absolute Percentage
Error (MAPE), this study aims to drive significant
progress in the automotive industry by optimising
inventory management reducing storage costs and
improving delivery efficiency to ensure smooth business
operations moreover the integration of visually engaging
dashboards for real-time analysis of projected values
plays a pivotal role in identifying crucial monthly demand
trends this integration not only enhances operational
efficiency but also fosters enriched customer engagement
thereby facilitating sustained advancement within the
automotive sector.
Keywords :
Time Series Analysis, Demand Forecasting, Inventory Management, Deep Learning, Prophet, Supply Chain.
