Neo Journal of Economy and Social Humanities (NEJESH) Volume 5.
Issue 1, 2026.
107-115 AU AU
ISSN: 2828-6480
http://dx.
org/10.
56403/nejesh.
Decision Support System for Digital Business Optimization:
A Comparative Analysis of Sales Forecasting Using the Moving Average Method Dely Indah Sari1.
I Made Sondra Wijaya2.
Muhammad Rizki3
1,2,3
Digital Business.
Institut Teknologi Batam.
Batam.
Indonesia.
Article Information Article History Received.
January 24, 2026 Revised.
February 5, 2026 Accepted.
February 13, 2026 Published.
February 14, 2026
_____________________
Corresponding Author:
Dely Indah Sari.
Digital Business.
Institut Teknologi Batam.
Batam.
Indonesia.
Email:
deli@iteba.
sondra@iteba.
ABSTRACT
This research aims to analyze and compare the accuracy of sales prediction fore-casting methods using three types of Moving Average (MA), namely Simple Moving Average (SMA).
Weighted Moving Average (WMA), and Trend Semi Average in the context of digital business optimization.
The MA method is used to identify sales trends and assist in planning more effective business strategies.
The data used in this research is historical sales data from digital business companies during a certain period.
Each MA method is applied to the data to produce sales predictions.
The accuracy of each method was evaluated using Mean Absolute Percentage Error (MAPE) as the main metric.
This research provides an important contribution for companies in choosing the most appropriate forecasting method to increase operational efficiency and effectiveness.
Recommendations resulting from this research can help companies optimize sales strategies and manage inventory better, to increase competitiveness in the market.
Keywords: forecasting, moving average, trend semiaverage, weighted moving average.
AU INTRODUCTION
In the current era of digital transformation, which continues to develop rapidly, it will change the way consumers interact and shop.
To be able to compete effectively, digital companies need to have a strong strategy to manage their business.
One important aspect of business management is the ability to predict sales accurately.
Reliable sales predictions enable companies to optimize inventory, manage budgets and plan effective marketing strategies.
This garment company focuses on certain types of clothing, such as casual clothing, sportswear, uniforms, or formal clothing, as well as various other textile products.
Characteristics of garment companies work with a variety of textile materials, such as cotton, polyester, silk, wool, and other synthetic materials.
The initial stage of the production process is where clothing models or textile products are designed.
Cutting means cutting fabric according to a design pattern to form parts of clothing.
Stitching is the process of combining fabric parts to form the final product.
Finishing includes adding details, washing, ironing, and packing.
In meeting the demand for garment products, companies with forecasting results, management can decide on the right policy in making decisions related to the company.
https://internationalpublisher.
id/journal/index.
php/Nejesh Figure 1.
Sales Diagram Company Forecasting methods or sales forecasting play an important role in helping businesses understand future trends based on historical data.
The moving average (MA) method is one of the popular techniques used in sales data analysis (Putramasi Hintarsyah et al.
, 2.
There are several uses of technology to support forecasting at company X using mathematical methods, namely Moving Average & Trend Semi Average.
Moving Average is a method of forecasting value alignment by taking a group of observed values AUand then finding the average, then using the average as a forecast for the next period.
The term moving average is used because every time new observational data becomes available, a new average figure is calculated and used as a forecast.
Trend Semi Average can be used for forecasting purposes by forming an equation such as regression analysis(Irawan et al.
, 2.
The semi average method can be used with even or odd amounts of data.
The trend equation obtained using a semi average can not only be used to determine the trend in the value of a variable from time to time but can also be used to predict the value of a variable at a certain time (Patandean et al.
, 2.
The importance of sales predictions in a company, in fact, by understanding sales patterns and trends, businesses can plan more targeted marketing, increase promotional effectiveness, and maximize return on investment from marketing activities.
From the background that has been created, the problem formulation can be described as follows what is the behavior and sales scheme based on the amount of production in the company.
In determining production targets for the coming year, how can company X identify forecasting based on sales in the previous year? What are the forecasting results obtained when using the Moving Average and Trend Semi Average methods.
This research aims to analyze and compare the effectiveness of the moving average method in predicting sales at a company (Zahri, 2.
The main goal of production forecasting is to ensure that the company can meet customer demand without experiencing excess or shortage of inventory (Adiningsih et al.
, 2.
Figure 2.
Flowchart Forecast Comparative Analysis Study of Sales Projections Using the Moving Average Method in Digital Business Optimization is an analysis method applied to predict future data based on existing This moving average method functions by calculating the average of the values AUof several previous periods to predict the values AUfor the next period.
The research design steps taken include determining the research objective, namely, to forecast sales, demand, production, or other variables (Eko siswanto et al.
, 2.
There is the first type of forecasting method, short-term forecasting, namely a period of less than 3 months, for example planning, purchasing materials, work scheduling and employee Second, medium term forecasting, namely a period of between 3 to 18 months, for example forecasting for sales planning, production planning and non-permanent workforce All three long-term forecasts cover greater than 18 months (Rachman, 2.
For example, capital investment planning, facilities planning and planning for R&D.
LITERATURE REVIEW
Influence of the product life cycle.
Another factor that must be considered when making sales forecasts, especially long-term sales forecasting, is the product life cycle, where sales of products and services do not occur at a constant level throughout their life, in fact almost all products successfully go through four stages, namely introduction, growth, maturity and decline (Yudaruddin, 2.
Benefits and Uses of Moving Average Trend Identification Helps in identifying long-term trends by averaging short-term fluctuations in data.
Noise Reduction: Reduces the AunoiseAy in time series data that is often caused by random fluctuations.
Prediction and Forecasting: Used in forecasting to predict future values AUbased on historical trends.
Technical Analysis in Finance in technical analysis, moving averages are used to identify buy or sell signals based on price Advantages and Disadvantages of Moving Averages.
The simple and easy to understand advantage of moving averages is that they are easy to calculate and implement.
for data with clear trends effective in averaging data and identifying underlying trends (Anjani et , 2.
Moving Averages are ideal for companies that need to adjust short-term strategies based on seasonal changes or rapidly changing data patterns.
Trend Semi Average is more suitable for long-term trend analysis that helps companies in strategic planning, such as expansion or capacity adjustments based on consistent growth predictions(Azahra et al.
, 2.
Forecasting is an important decision in a company where the company can determine the level of production that needs to be prepared for the future.
Determining the level of production is a supply that is influenced by the amount of market demand that the company can influence (Dua, 2.
To help achieve an optimal decision, appropriate, systematic, and accountable methods are One of the tools needed by management and an integral part of the decision-making process is the forecasting method.
RESEARCH METHOD
Moving Average (MA) is a statistical method used to analyze time data .
ime serie.
with the aim of identifying long-term trends and smoothing short-term fluctuations in the data .
Moving Average calculates the average of a particular subset of data in a data series, thereby helping to reduce the effects of random variation and making underlying trends more visible .
The moving average method is an analysis technique used to even out data fluctuations to identify clearer trends.
This technique is often used in time series analysis and forecasting.
Moving Average (MA) is a statistical method used to analyze data in the form of a time This method averages data by calculating the average of a certain number of data points and moving from one point to the next in the time series.
There are several types of moving averages that are commonly used.
There are several types of Moving Averages that are often used in data analysis and forecasting:
ycAya = ycUycOe1 ycUycOe2 ycUycOe3 ycu Where:
n= number of periods in the moving average t= demand for period i X= data value Moving average data is usually interpreted as follows uptrend if the moving average tends to increase, this indicates an upward trend in the original data.
Downtrend if the moving average tends to decrease, this indicates a downward trend in the original data.
Stable if the moving average moves horizontally, it means the original data does not show significant changes.
Weighted Moving Average (WMA) assign different weights to each period of data, with newer data usually given a higher weight.
ycOycAyayc = ) ( .
cUycycuycO.
ycUycOe1ycuycO2 ycUycOe2ycuycO1 yc1 yc2 yc3 Where:
Xt = sales value for month t w1, w2, w3 = weight of each month Trend Semi Average is a technique in time series analysis that is used to estimate or determine the direction of data trends.
This method involves dividing the data into two equal segments, then calculating the average of each segment to obtain a picture of the pattern.
ycIyceycaycuycuycc ycaycyceycycayciyceOeyceycnycycyc ycaycyceycycayciyce yca = yaycnyceyceyceycyceycuycayce ycnycu ycyceycaycyc ycayceycycyceyceycu yciycycuycycyyc .
Calculate the intercept value .
using one of the group midpoints.
For example, let's take the midpoint of the first group with X=0 as the reference point for this year:
ycU = yca ycaycU Benefits and Uses of Moving Average Trend Identification Helps in identifying long-term trends by averaging short-term fluctuations in data.
Noise Reduction: Reduces the AunoiseAy in time series data that is often caused by random fluctuations.
Prediction and Forecasting: Used in forecasting to predict future values AUbased on historical trends.
Technical Analysis in Finance: In technical analysis, moving averages are used to identify buy or sell signals based on price Advantages and Disadvantages of Moving Averages.
The simple and easy to understand advantage of moving averages is that they are easy to calculate and implement.
for data with clear trends effective in averaging data and identifying underlying trends.
RESULTS AND ANALYSIS
Moving average (MA) helps minimize the effects of short-term fluctuations in the Table 1 Moving Average Month Sale (Uni.
Jan
Feb
Maret
April
Mei
Jun
Moving Average ycAya ycAyceycn = 750 900 1200 1450 1600
ycAya yaycycuycn = 900 1200 1450 1600 1550
Jul
1600 1550 1850 2000
= 1.
912, 5
1550 1850 2000 2485
= 1.
971, 25
ycAya ycCycoycycuycayceyc = 1450 1600 1550 1850 = 1.
ycAya ycCycoycycuycayceyc = Des = 1.
ycAya ycCycoycycuycayceyc = Nov = 1.
487, 5
1450 1600 1550 1500
ycAya ycIyceycyycyceycoycayceyc = Okt = 1.
ycAya yayciycycycycyc = Sep
1850 2000 2485 2760
= 2.
273, 75
Based on the 4 Month Moving Average calculation, estimated sales for January next year are 2,274 units.
The 4 Month Moving Average shows that product demand tends to increase over Figure 3.
Moving Average Figure 7.
The company experienced a slow increase in sales throughout the year.
applying a 4-month period, any possible seasonal variations in sales have been smoothed out.
This helps companies understand a more consistent long-term demand picture.
Based on these findings, the company can prepare the right inventory or production to meet the estimated need of around 2,274 units in January next year.
If the trend remains upward, companies may have to think about strategies to increase production or logistics capacity to meet the surge in demand in the coming Chart Analysis there is a visible improvement that there is an overall sales growth trend in the last few months.
Softer changes in the moving average line help reduce the impact of significant monthly fluctuations, making the upward trend more obvious.
Moving Averages like this are useful for managers to forecast trends in the following month and make more accurate inventory decisions.
Trend Semi Average is a trend analysis method used in time series analysis to predict gradual changes in data patterns over time.
In this method, the data is divided into two equal parts, then the average of each part is calculated.
This average value is then used to describe trends in the data, which makes it easier to identify the general direction of the data, whether it is increasing or This method is simple but effective, especially when the data shows a consistent linear pattern or general trend.
It is usually used to provide an initial picture of medium-term trends, making it suitable in business or economic analyzes that observe gradual changes in sales data, production, or other economic indicators.
Table 1.
Trend Semi Average Years Sale (Uni.
Trend Semi Average Average data 1 .
6 Ae 2.
1550 2340 2450 2100 yaycyceycycayciyce = = 4 = 2110 Average data 1 .
0 Ae 2.
2500 2700 3000 3500 yaycyceycycayciyce = = 2925 Calculating Trend Slope, b:
The trend slope .
is calculated by subtracting the second group mean from the first group mean, then dividing by the number of years between the two midpoints of the two groups .
years in betwee.
ycIyceycaycuycuycc ycaycyceycycayciyceOeyceycnycycyc ycaycyceycycayciyce yca = yaycnyceyceyceycyceycuycayce ycnycu ycyceycaycyc ycayceycycyceyceycu yciycycuycycyyc
yca= 2925Oe2110
= 203, 75
Slope b= 203.
In 2017 .
he midpoint of the first grou.
, the average sales value was 2340AU AU 2340=a 203.
75 x 0 Then a=2340 Equation Y=2340 203.
Sales Predictions for 2024:
For 2024, we calculate the value of X as the difference in years from the base year .
X= 2024 Ae 2017 = 7 Y= 2340 203.
= 2340 203.
= 2340 1426.
= 3,766.
Sales prediction results for 2024 using the Trend Semi Average method are 3,766.
thousand units.
Figure 4.
Trend Semi Average AU Weighted Moving Average (WMA) is a data analysis technique that pays more attention to the most recent data when calculating the moving average, making it more responsive to recent changes in the data when compared to the Simple Moving Average (SMA).
On the WMA, more recent information is rated as more relevant and has a greater impact on the mean than older Figure 5.
Weighted Moving Average (WMA) AU This method is very suitable for application in time series data analysis, especially in sectors such as sales, production, finance, and stock exchanges.
Table 2.
Weighted Moving Average (WMA) Month Sale (Unit.
Weighted Moving Average (WMA) ycOycAyayc = .
9 8 7
16650 12000 11515
ycOycAyayc = = 1673
10 9 8
20000 16650 11515
ycOycAyayc
= 1783
11 10 9
27335 20000 16650
ycOycAyayc = = 2132 ycOycAyayc = ycOycAyayc = = 1562 ycOycAyayc = ycOycAyayc = .
6 7 8
12000 11515 9300
ycOycAyayc ycOycAyayc = ycOycAyayc = = 1600 ycOycAyayc = ycOycAyayc = .
7 6 5
11515 9300 8000
ycOycAyayc = ycOycAyayc = ycOycAyayc = .
6 5 4
9300 8000 5800
ycOycAyayc = = 1540 ycOycAyayc = ycOycAyayc = .
3 4 5
3600 5800 8000
ycOycAyayc = = 1150 ycOycAyayc = ycOycAyayc = .
4 3 2
5800 3600 1800
ycOycAyayc = = 1244 ycOycAyayc = ycOycAyayc = .
1 2 3
750 1800 3600
= 1025
ycOycAyayc = .
12 11 10
33120 2760 20000
ycOycAyayc = = 1693 ycOycAyayc = ycOycAyayc = CONCLUSION Moving Average calculates the average value over a certain time period by periodically replacing old data with new data.
The goal is to smooth data fluctuations to identify long-term Trend Average: Use historical data to identify long-term patterns .
in data.
Typically utilize linear regression or similar methods to predict business direction.
Weighted Moving Average (WMA) This method gives greater weight to data that is considered more important or relevant.
The weight can be adjusted based on importance or priority.
The choice of method depends on the type of data, the purpose of the analysis, and the desired level of sensitivity to change.
For a more stable and simple analysis.
Moving Average is more suitable.
you need a long-term view, use Trend Average.
If you want to focus on a specific aspect of the data.
Weighted Average provides greater flexibility.
In business practice, a combination of methods is often used to obtain a more comprehensive analysis.
ACKNOWLEDGEMENTS
The authors would like to express their sincere gratitude to the LPPM ITEBA for the support, guidance, and facilitation provided throughout the completion of this research entitled Decision Support System for Digital Business Optimization: A Comparative Analysis of Sales Forecasting Using the Moving Average Method.
The support from LPPM ITEBA has been invaluable in enabling this study to be carried out successfully.
REFERENCES