Online MBA Programmes Assignment Sample

1.0 Introduction of Online Mba Programmes Assignment

Dunelm is a large home furnishings retailer of bedding, curtains, and home accessories based in the United Kingdom today’s competitive retail landscape requires precise sales forecasting to manage projects effectively and make sound business decisions. The purpose of this report is to show three numerical forecasting techniques Moving Average, Exponential Smoothing, and Trend Forecasting to depict future sales projections of Dunelm’s product line. In doing so, future sales patterns are to be identified which should be useful in making important decisions concerning inventory management, marketing, and resource allocation strategies. Through the strengths and weaknesses of these techniques will enable Dunelm to improve its operations and provide better solutions for consumer demand needs. This study will also demonstrate how the various forecasting techniques used can help improve business outcomes and achieve organizational objectives in the highly competitive retail environment.

24/7 support, expert writers, and fast delivery New Assignment Help is your best choice for trusted Assignment Help in the UK.

2.0 Overview of Numerical Forecasting Methods

Moving Average

The moving average is another form of simple forecasting technique that eliminates short-term cyclical variations but retains the trend. It is calculated for a fixed number of the most recent data values, usually periods like 3 months or 6 months. As more data values are accumulated, older data values are deleted so that the forecast can be updated continuously. In retail, it is useful in making an average that erases the daily fluctuations in the sales so one can see what the trend of the sales looks like, it is also helpful in trading since they do not have high sensitivity in volatile or trending markets (Zunic et al. 2020). Moving averages while helping clear up data are slow to adjust to change and this makes them somewhat less sensitive to the high volatility environment like the retail.

Exponential Smoothing

This is one of the most common techniques that involve assigning different weights to different values in the sequence. By assigning progressively lower weights to the older values and providing higher weight to the latest values this method is effective for seasonal and other data that exhibit short-term fluctuations but should be more sensitive to the changes in the trends as they are developing. This allows the forecaster to control the magnitude of the responsiveness to the recent change by selecting the characteristic smoothing parameter α (Shillong, 2021).Retail sales are forecasted using exponential smoothing because it is more effective as it factors the recent market trends and provides a stable forecast value. It also might fail when facing highly volatile data and might require some modification like a trend or seasonal smoothing to identify such trends.

Trend Forecasting

Trend estimation is one of the methods of analysis used to determine long-term trends in time series data sets. It is built based on linear regression; it analyses line progression trying to predict further figures. In retail trend forecasting assists in determining whether there are improvements, deterioration, or business as usually into the future. It is most helpful for business planning because it allows for the coordination of inventory and marketing initiatives with expectations for future demand. However this method is divided with the notion that one tends to continue to move in the direction of its previous trends, especially in volatile markets (Niu, 2020). As it have been seen, the trend forecasting serves long-term expectations well because sudden short-term fluctuations are not easily detected.

3.0 Data Collection and SPSS Analysis

Data Description

The source of the sales data, data for this project is sourced from the publicly available Retail Sales Dataset on Kaggle. This dataset contains retail sales figures, which is suitable for time series forecasting. The dataset contains records of retail sales transactions for the year 2023, where the total features are 11 which indicates different transactions. The data for each record includes Transaction ID, Date, Customer ID, Gender, age, product category, quantity, price per unit in pounds (£), and total amount in pounds (£). The transactions include products such as Beauty, Clothing, and Electronics among others with unit prices as low as £25 and extending up to £500. Biologically, our customer base comprises both male and female customers who are aged between 23 and 63 years. The sales data is collected based on months in a year for the year 2023 where the data starts from January and ends in December which is good for time series.

Data Analysis

Moving Average

Figure 1: Model description and Model fit

In this image, there is a table with the identification of the Model Description and two columns. Across the top row, the first column provides the labels Model ID and Total Sales Amount (€), whilst the second column displays Model_1.The chart model type is mentioned as ARIMA (0,0,4). This appears to be an Overview of total amounts from the sales time series most probably an ARIMA model. The Model Fit table in this figure shows several fit statistics applicable to a statistical model. Many of them are R-squared, RMSE, MAPE, and the other that is calculated at percentiles of data distribution for this or that case. Indeed, the fact that the values at different percentiles are similar may point to the fact that this is built around one model fit rather than across multiple model fits (Wisesa et al. 2020). As indicated in the table below, the model offers precise details on the performance and accuracy of the model. The number of predictors is zero.

Figure 2: Model statistics

This table represents a bar-formatted Model Statistics table that contains values of metrics for a statistical model. This suggests no predictors were entered, an R-squared value of 0.017, and a Q statistic of 25.334 for 18 lag intervals with 16 degrees-of-freedom P <.064. Outliers = 0 the table also notes that there was no indication of any outlier that was observed (Pan and Zhou, 2020). During this discussion, an overview of the fit and diagnostic statistics on a model in time series has been presented.

Exponential Smoothing

Figure 3: Model description and Model fit

This image has two details of the dataset: a “Model Description” table and a “Model Fit” table. The model type is "Brown". The fit statistics table gives other values of different percentiles such as R squared, RMSE, MAPE, and so on. For example, the Stationary R-squared equals 0. 801 for all percentiles, whereas the R-squared equals -0. 008 ( Kolkova, 2020). Even other measures such as the RMSE (562.205) and MAPE (417.064) are apparently identical for all percentile values. This indicates, therefore, that a time series model with performance measures that are described as stable may be appropriate for the case.

Figure 4: Model statistics

The Model Statistics table below presents some of the features of the statistical model used in the study. It proves that no predictors were included in the model. The Stationary R-squared is 0.801 as it was with the previous image. Ljung-Box Q (18) statistic is equal to 44.220 at 17 degrees of freedom less than 0.001 showing that there might be an issue with autocorrelation in the residuals (Abolghasemi et al. 2020). This table holds the basic summary information about the model fit and diagnostics of time series analysis.

Trend Forecasting

Figure 5: Model summery

The other model statistics of interest are presented in the Model Summary table. The coefficients show that, in the present model, the R-value is 0.002, R square 0.000, and the adjusted R square -0.001, this makes the current model, explain a very negligible variation of the total variance of the dependent variable total sales amount. The Standard Error of the Estimate = 560.277, while the Durbin-Watson statistic = 1.801 (Fildes and Goodwin, 2021). The indicators are shown in the footnotes, also asserting that Date is the constant predictor and the dependent variable. This summary [of other statistics] also indicates a rather unsuitable fit of the model.

4.0 Evaluation of Forecasting Methods

Strengths

• Moving Average: This technique is easy to apply and beneficial in eliminating short-run volatility thus useful in detecting longer-run trends of sales. This is particularly the case when there is market stability so that the sales experience little fluctuations over time.
• Exponential Smoothing: this is more important to recent data and, thus is more sensitive to short term pattern. It also is effective in volatile situate that exhibit fluctuation in demand over a short period of time, thus, produces improved short-run sales forecasting.
• Trend Forecasting: Through regression, trend forecasting is very suitable for detecting long-term sales trends meaning that businesses can be able to predict how the market will shape up in the future based on how it has been in the past (Efat et al. 2024). It is particularly effective in conditions, where there are regular fluctuations in growth and decline trends over a given time.

Limitations

• Moving Average: The main weakness of the method is that it has a lag effect and that it is not immediately responsive to fluctuations in sales, it is therefore not suitable for volatile markets. Moreover, the time interval has to be selected, which is often a matter of judgment.
• Exponential Smoothing: It is sensitive to the current changes, but it may be highly sensitive to short-term signals; therefore, it might not be very valuable in long-term predictions. It also presupposes the proper selection of the smoothing constant (α) for striking a proper equilibrium between responsiveness and stability.
• Trend Forecasting: The assumption of this method is based on the history which is not accurate in the current and highly competitive markets (Ensafi et al. 2022). It is less fit for use when making short-term forecasts and might result in overlooking a big shift in consumer behavior.

5.0 Application in Project Management Context

Dunelm uses sales forecasts to improve decision-making

A sales forecast is useful to Dunelm in matters of production, procurement, and the preparation of its budgets. In being able to predict future sales, the company can regulate resource use, prevent overstocking and untimely stockouts, and improve its overall performance.

The implications for inventory management, marketing strategies, and resource allocation

It suggests the right quantity of stock to purchase to meet demand, directs the appropriate marketing activities based on demand patterns, and allocates resources to various departments effectively (Weng et al. 2020). This enables the acquisition of the right products at the right time that meet the customer needs hence reducing on cost of excess inventories.

The significance of accurate forecasts in achieving business objectives

It is imperative for Dunelm to forecast its sales because the goal of the business is to increase its profits, increase its customer’s satisfaction levels, increase its competitiveness among other businesses. In turn, accurate projections help to improve the financial planning and management, overall expenses, and the ways that firm can achieve growth to meet the challenges and demands of the marketplace.

6.0 Prediction of Future Trends

According to the results of Moving Average, Exponential Smoothing, and Trend Forecasting, the sales of home furnishing products in the future remain stable but not very intensive from the side of the business The results also show the seasonal changes in the home furnishing industry are high during holidays to present bedding products and home accessories are more demanded. Enhanced focus on the latest data shows that any short-term change, such as a decrease in demand after the holiday sales season, will significantly affect the forecasts (Zhang et al. 2020). The trend also predicts a steady increase in home renovation and, therefore, sales spending. This enables Dunelm to understand when demand will be higher during upcoming holidays, and prepare for the necessary inventory stocking and marketing campaigns.

7.0 Conclusion

In this report, the numerical forecasting techniques Moving Average, Exponential Smoothing, and Trend Forecasting have been computed on Dunelm’s sales datasets. Both approaches offered benefits with respect to predictive excellence for the future, Sales; Applications in stock control, marketing and resource use. Through forecasting, Dunelm can manage its operations effectively, and thus accomplish the corporate goals and objectives required when operating in a competitively evolve retail environment. These is also supports strategic decision making which is helpful for the company to allocate resources effectively.

Reference List

Journal

• Abolghasemi, M., Hurley, J., Eshragh, A. and Fahimnia, B., 2020. Demand forecasting in the presence of systematic events: Cases in capturing sales promotions. International Journal of Production Economics, 230, p.107892.
• Efat, M.I.A., Hajek, P., Abedin, M.Z., Azad, R.U., Jaber, M.A., Aditya, S. and Hassan, M.K., 2024. Deep-learning model using hybrid adaptive trend estimated series for modelling and forecasting sales. Annals of Operations Research, 339(1), pp.297-328.
• Ensafi, Y., Amin, S.H., Zhang, G. and Shah, B., 2022. Time-series forecasting of seasonal items sales using machine learning–A comparative analysis. International Journal of Information Management Data Insights, 2(1), p.100058.
• Fildes, R. and Goodwin, P., 2021. Stability in the inefficient use of forecasting systems: A case study in a supply chain company. International Journal of Forecasting, 37(2), pp.1031-1046.
• Kolkova, A., 2020. The application of forecasting sales of services to increase business competitiveness. Journal of Competitiveness, 12(2), p.90.
• Niu, Y., 2020, October. Walmart sales forecasting using xgboost algorithm and feature engineering. In 2020 International Conference on Big Data & Artificial Intelligence & Software Engineering (ICBASE) (pp. 458-461). IEEE.
• Pan, H. and Zhou, H., 2020. Study on convolutional neural network and its application in data mining and sales forecasting for E-commerce. Electronic Commerce Research, 20(2), pp.297-320.
• Shilong, Z., 2021, January. Machine learning model for sales forecasting by using XGBoost. In 2021 IEEE international conference on consumer electronics and computer engineering (ICCECE) (pp. 480-483). IEEE.
• Weng, T., Liu, W. and Xiao, J., 2020. Supply chain sales forecasting based on lightGBM and LSTM combination model. Industrial Management & Data Systems, 120(2), pp.265-279.
• Wisesa, O., Adriansyah, A. and Khalaf, O.I., 2020, September. Prediction analysis sales for corporate services telecommunications company using gradient boost algorithm. In 2020 2nd International Conference on Broadband Communications, Wireless Sensors and Powering (BCWSP) (pp. 101-106). IEEE.
• Zhang, C., Tian, Y.X., Fan, Z.P., Liu, Y. and Fan, L.W., 2020. Product sales forecasting using macroeconomic indicators and online reviews: a method combining prospect theory and sentiment analysis. Soft Computing, 24, pp.6213-6226.
• Zunic, E., Korjenic, K., Hodzic, K. and Donko, D., 2020. Application of facebook's prophet algorithm for successful sales forecasting based on real-world data. arXiv preprint arXiv:2005.07575.

Author Bio

Cameron Lee   4 years | MSc in Management

I am Cameron Lee and I have completed an MSc in management from a renowned University of the United Kingdom. For the last 4+ years, I have been assisting students with their assignments, dissertations, essays and other academic papers. If you are struggling with your management task you can have me by your side for help. I would like to give you the best results.