Using the Arima Method with Minitab Applications for Forecasting Work Order Projects of Casting Construction (Case Study: PT.Bumi Sarana Beton)

The ARIMA method is a non-stationary homogeneous time series model that uses the procedure for applying the Autoregressive model or scheme and the Moving Average in preparing its forecasts. The purpose of this study was to determine the application of the Autoregressive Integrated Moving Average (ARIMA) method and minitab application at PT. Bumi Sarana Beton in planning the estimated number of work orders. It can be concluded that for K-225 the best model is the ARIMA model (2,0,0) because it has the lowest MSE value and for K-400 the best model to use is the ARIMA model (2,0,2 ) because it has the smallest value. From the results of the research that has been done, it can be concluded that the production of K-225 in 2022 is 3573.50 m3, while using the ARIMA method in 2023 the total production is 3920.61 m3 and in 2024 the total production is 3573.50 m3. to 3824.36 m3. Production of K-400 in 2022 was 2015.00 m3, while using the ARIMA method in 2023 a total production of 1857.07 m3 was obtained and in 2024 a total production of 2045.89 m3 was obtained.


Introduction
Forecasting is one of the statistical methods that play an important role in decision making.Forecasting serves to estimate what will happen in the future based on past data.[1] The ARIMA method is a nonstationary homogeneity time series model that uses procedures for applying Autoregressive and Moving Average models or schemes in the preparation of forecasts.[2] Conducting research with the title Analysis and Implementation of Data Mining to Predict the Number of Building Materials Using the Autoreggresive Intergrated Moving Average (ARIMA) Algorithm.[3]  Work Orders Using the ARIMA (Autoregressive Integrated Moving Average) Method.[4] Research on Material Inventory Study Based on Material Requirement Planning (MRP) Method on Fresh Concrete Production (Case Study: PT.Bumi Sarana Beton).[5] Research on the Use of Arima Method with Minitab Application for Forecasting Revenue Data of Putra Mndiri Stone Breaking Company.[6] The results showed that PT Bumi Sarana Beton Makassar (Hadji Kalla project) has implemented responsibility accounting.[7] Based on the analysis of PT Surindo Andalan data from the author regarding forecasting with the ARIMA approach in the MINITAB program, it can be concluded that the best ARIMA model with the smallest mean square of 2709.7 and the smallest mean square error of 225.808333 is the ARIMA (2,0,4) model.[8] Project scheduling using precedent diagrams and project equipment transfer matrices can be used as an alternative to scheduling projects simultaneously.[9] Using the MRP method with lot sizing calculations at the Masohi Power Center can help plan the raw material requirements for each product item in a timely manner, so that the production process will be carried out properly according to the planned capacity and orders can be fulfilled on time.[10] Implementing a project is the process of converting inputs in the form of activities and resources into outputs as specified.[11].

Research Method
The data used in this research is secondary data.The author obtains data by receiving work order recapitulation data for construction projects from January 2022 to December 2022.
This research used MINITAB is a computer program designed to perform statistical processing (Sunday, n.d.).Minitab combines ease of use like Microsoft Excel with the ability to perform complex statistical analysis (Simarmata, 2010).
MINITAB is statistical software that provides a wide range of capabilities for both basic and advanced statistical analysis.The program's powerful capabilities and ease of use make it ideal as a teaching tool.Minitab was developed at Pennsylvania State University by researchers Barbara F. Ryan, Thomas A. Ryan, Jr., and Brian L. Joiner in 1972.MINITAB has become the standard for statistical learning.And since MINITAB is the leading package used to improve processes and quality within companies, students studying MINITAB definitely benefit from knowing and also being able to use the tools used in the real world of business (Ryan, Joiner, & Cryer, 2005).

Description And Technical
1) The steps for analyzing data using the ARIMA model are as follows: 1. Displaying data plots to find out whether the data is patterned or not 2. Identify stationary or non-stationary models.
3. Estimation of parameters by conducting hypothesis testing to determine whether or not the parameters are significant.Hypothesis: H0 = parameter that is not significant H1 = significant parameter Tolerance levels (α): 5% = 0.05 2) Research steps using the minitab19 application The general stages in the analysis carried out are as follows.
1. Model identification Model identification is used to select the right model for forecasting, the stages in the identification are: a. Create time series plots The purpose of making this plot is to find out the trend of a data.
b. Make ACF and PACF plots The next step is to make ACF and PACF plots to find out the stationarity of the variance and average estimates of the model to be used.

c. Data Stationarity Test
After analyzing the data by looking at the time series plots as well as the ACF and PACF, the data stationarity is checked.A data is said to be stationary in terms of variance if it has a Rounded Value of 1, while a data is said to be stationary in average if the trend graph in the trend analysis plot is horizontal, which means that the average and trend of the data do not change over time.If the data used in the analysis turns out to be non-stationary, it is necessary to carry out stationarity by transforming the data on data that is not stationary in terms of variance and differentiation on data that is not stationary in terms of average.In this case, the first is to perform a transformation if the data is not stationary in the new range, then the differentiation if the data is not stationary in the average.

Estimation or Estimated Model
If we have done the analysis from the previous step, we get the equation for the forecasting model.The next step is to estimate the model to determine whether the model is feasible or not for forecasting.The steps are by looking at the p-value of Box-Cox and also AR and MA where each value cannot be more than 0.05.

Verify
After obtaining the appropriate equation model for use in forecasting, the next step is to verify the best equation model for use in forecasting.In this case it can be searched by looking at the smallest error or MSE.

Forecasting or Forecasting
Forecasting is an activity of estimating what will happen in the future.If observed from the results of the two data plots of K-225 and K-400 production above, it can be seen that the K-225 production has a slightly decreasing graph and then rises again over time, the K-400 production can also be seen to have a graph initially very high until it decreases and slowly rises and falls over time.From the data plot above, it can be seen that the data that has been transformed is still not stationary on average because the trend line on the graph is getting higher which means that the average is not fixed, the solution is to do differentiation.

Source: research of data analysis in minitab 19
After obtaining the differentiation results, it is checked again whether it is stationary on average by making a trend graph, following the output results.From the output results above, it can be concluded that the data is stationary on average because the trend line on the graph is horizontal, which means that the average value of the data is fixed.In the plot above, it can be seen that both plots for K-225 and K-400 data do not cross the red significant line, therefore the data is stationary on average.   in the first year are unstable where each month there is a decrease and increase.The highest monthly production rate in May was 291,446 m3 and the lowest was in January at 39,473 m3.For the second year, it experienced the same thing in the first year with the highest production rate in January at 287,841 m3 and the lowest in May at 46,618 m3.

Conclusions And Sugesstion 5.1 Conclusions
1. Based on the results of model identification, at least 3 models are obtained that allow it to be used in forecasting, namely ARIMA (2,0,0), ARIMA (2,0,2) and ARIMA (0,0,2) for K-225 and K-400 respectively, but after checking again with the model estimation there are only 2 models that are suitable to be used as forecasting equation models, namely ARIMA (2,0,0) and ARIMA (0,0,2) for K-225 while K-400 is ARIMA (2,0,2) and (0,0,2 has the smallest MSE value as well for the K-225 and K-400 production forecasting model at PT. Bumi Sarana Beton.2. From the results of the research that has been done, it can be concluded that the production of K-225 in 2022 is 3573.50m3 while when using the ARIMA method in 2023 the total production is 3920.61m3 and in 2024 the total production is 3824.36m3.K-400 production that in 2022 a total of 2015.00 m3 was obtained, while when using the ARIMA method in 2023 a total production of 1857.07 m3 was obtained and in 2024 a total production of 2045.89m3 was obtained.

Suggestion
1. From the results of the research is expected to add input to the company PT.Bumi Sarana Beton so that later can be as a reference in improving the quality of the company's production.2. When viewed from the discussion of the previous chapter that ARIMA has shortcomings in its forecasting which is quite good in short-term forecasting and for long-term forecasting it has an increasingly large error, and this is what needs to be refined.

Figure 3 .
Figure 3. Trend plot of K-225 production yield analysis

Figure 4 .
Figure 4. Trend plot of K-400 production yield analysis

Figure 10 .
Figure 10.Plot trend analysis of K-400 transformation data

Figure 11 .
Figure 11.Trend plot of K-225 differentiation result data

Figure 12 .
Figure 12.Trend plot of K-400 differentiation result data

Figure 13 .
Figure 13.ACF plot of TRANS 1 result data for K-225

Table 1 .
Differentiation result 1 of K-225 transformation data Source: research of data analysis in minitab 19 129 Using the Arima Method with Minitab Applications for Forecasting Work Order Projects of Casting Construction (Case Study: PT.Bumi Sarana Beton)

Table 1 .
Differentiation result 1 of K-400 transformation data

Table 1 .
1 K-225 production forecasting results Using the Arima Method with Minitab Applications for Forecasting Work Order Projects of Casting Construction (Case Study: PT.Bumi Sarana Beton)

Table 1 .
2 K-400 production forecasting resultsSource: research of data analysis in minitab 191.Based on the forecasting results in table 1.1, it is found that the production results for K-225 in the first year are unstable where each month there is an increase and decrease.The highest monthly production figure in February was 395,289 m3 and the lowest was in May at 275,324 m3.For the second year experienced the same thing in the first year with the highest production figure in March at 324,449 m3 and the lowest was in June at 314,668 m3. 2. Based on the forecasting results in table 1.2, it is found that the production results for K-400