Mean Absolute Percentage Error Values Research Articles

Peramalan Produksi Kelapa Sawit dan Karet di Provinsi Kalimantan Selatan

In Indonesia, the plantation sub-sector has an important role in increasing state revenue through the exports of its products, besides the mining and gas sector. The most widely produced plantation crops in Indonesia are oil palm and rubber and South Kalimantan is one of the top 10 provinces in Indonesia with oil palm plantations. This study aims to detect the correct forecasting model for data on oil palm crops and rubber production in South Kalimantan Province and to analyse the forecasting results for oil palm crops and rubber in South Kalimantan Province using the double exponential smoothing method. This research was conducted for 8 months (March 2022 to December 2022), using observational data from 2001 to 2021. Double Exponential Smoothing Holt was used in this study by looking at the error value obtained with the smallest Mean Absolute Percentage Error (MAPE). For palm oil production, the parameters α=0.8 and β=0.6 were the best parameters with a MAPE value of 8.05% and resulted in the forecasting of oil palm crops production in 2022 not increasing, in 2023 and 2024 experiencing an increase of 1%. As for forecasting rubber production, the parameters α=0.9 and β=0.9 are the best parameters with a MAPE value of 5.45% and forecasting rubber production in 2022 will increase by 1%, in 2023 and 2024 by 2%.

Mechanical behaviour of E-waste aggregate concrete using a novel machine learning algorithm: Multi expression programming (MEP)

Technological advancement encourages the usage of electronic appliances in daily life and makes it possible for users to switch to more advanced devices very easily and at a reasonable cost. As new devices are produced and manufactured at an alarming rate around the world, outdated old devices become e-waste. This research work aims at using a popular machine learning (ML) method known as Multi-expression programming (MEP) to examine the compressive strength (CS) and tensile strength (TS) of E-waste aggregate-based concrete (EWAC). 279 and 105 scientific entries for CS and TS, respectively, were culled from reputable literature. The ten convincing input parameters selected based on multicollinearity analysis (correlation matrix and variance inflation factor) were E-waste coarse aggregate (ECA%), E-waste fine aggregate (EFA%), water-cement ratio (w/c), age of concrete (A in days), fine aggregate water-absorption (WAF%), coarse aggregate water-absorption (WAC%), E-waste aggregate water-absorption (WAE%), E-waste aggregate specific-gravity (SGE), coarse aggregate specific-gravity (SGC), and fine aggregate specific-gravity (SGF). To estimate the functioning of the projected models, root-squared-error (RSE), mean-absolute error (MAE), mean-absolute-percent error (MAPE), Nash-Sutcliffe-efficiency (NSE), root-mean-squared error (RMSE), objective-function (OF), coefficient-of-correlation (R), root-mean-squared-logarithmic error (RMSLE), and performance-index (PI) were used. The R-value for both MEP models exceeds 0.9, showing “excellent” with MAPE values in the testing stage equals to 6.68% and 6.78% for the CS-MEP and TS-MEP models, respectively. While for non-linear regression (NLR) models, the MAPE exceeds 20% and 30%, respectively, making them unsuitable for future prediction. Moreover, the sensitivity analysis carried out to evaluate the MEP equations' consistency with the observed physical phenomena, indicates that for both CS and TS, the w/c, ECA%, and EFA% remain the most sensitive parameters with a sensitivity index greater than 0.60. Due to the accuracy and viability of developed models, they can be used to reduce the time needed for laborious laboratory tests.

Sales forecasting of marketing using adaptive response rate single exponential smoothing algorithm

Micro, small and medium enterprises (UMKM) is one of the important aspects to support the improvement of the economy in Indonesia. Zee Mart’s business is one of the UMKM shop in Pematang Siantar City with sales and purchase transaction activities for supplies. The purpose of this study is to predict the sales of Zee Mart store goods in the coming month using the adaptive response rate single exponential smoothing (ARRSES) method. ARRSES is a method with the advantage of having two parameters, alpha and beta, where alpha will change every period when the data pattern changes. The dataset obtained will be pre-processed through data selection, cleaning, and transformation. The best beta is determined based on the level of accuracy calculated using the mean absolute percentage error (MAPE). Model development using the ARRSES method will produce forecasting percentages and errors for each product using MAPE. The number of sales data is 23,092 before preprocessing and 23,021 after pre-processing, with the total quantity of goods sold being 149,764 of 1,492 products. The results obtained using sales data 23,021 show the lowest MAPE value of 9.85 at the best beta of 0.6 with the highest accuracy of 90.15% and the model is implemented into a web interface.

Deep learning approaches for MIMO time-series analysis

This study presents a comparative analysis of various deep learning (DL) methods for multi-input and multi-output (MIMO) time-series forecasting of stock prices. The analysis is conducted on a dataset comprising the stock price of Bitcoin. The dataset consists of 2950 rows from December 2017 to December 2021. This study aims to evaluate the performance of multiple DL methods, including Multilayer Perceptron (MLP), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Long Short Term Memory (LSTM), Bidirectional LSTM (Bi-LSTM), and Gated Recurrent Unit (GRU). The evaluation criteria for selecting the best-performing methods in this research are based on two performance metrics: Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE). These metrics were chosen for specific reasons related to assessing the accuracy and reliability of the forecasting models. MAPE is used to assess accuracy, while RMSE helps detect outliers in the system. Results show that the LSTM method achieves the best performance, outperforming other methods with an average MAPE value of 8.73% and Bi-LSTM has the best average RMSE value of 0.02216. The findings of this study have practical implications for time-series forecasting in the field of stock trading. The superior performance of LSTM highlights its potential as a reliable method for accurately predicting stock prices. The Bi-LSTM model's ability to detect outliers can aid in identifying abnormal stock market behavior. In summary, this research provides insights into the performance of various DL models of MIMO for stock price forecasting. The results contribute to the field of time-series forecasting and offer valuable guidance for decision-making in stock trading by identifying the most effective methods for predicting stock prices accurately and detecting unusual market behavior.

Machine learning approaches for forecasting inflation: empirical evidence from Sri Lanka

The aim of this study is to forecast the inflation rate using supervised machine learning models (SMLM). While SMLMs are widely used in various fields, they have also been widely applied to forecast inflation rates. Therefore, the main objective of this study is to identify the best model for forecasting inflation among four different SMLMs: LASSO regression (LR), Bayesian Ridge Regression (BRR), Support Vector Machine Regression (SVR), and Random Forest Regression (RFR) models. To achieve this objective, two different types of cross-validation techniques were used: K-fold cross-validation method (KCV) and walk forward validation ( ) methods. These techniques were used to estimate the parameters and hyper-parameters for each machine learning model with the aid of root mean square error. The mean absolute percentage error (MAPE) was used to compare the performance of the different SMLMs. Empirical evidence from Sri Lanka between 1988 and 2021 was used to test the performance of the SMLMs in forecasting inflation rates. The results show that the LR model with walk forward validation is the best method for forecasting the future inflation rate of Sri Lanka based on the MAPE value. Overall, this study demonstrates the effectiveness of SMLMs in forecasting inflation rates under the critical conditions and highlights the importance of employing appropriate cross-validation techniques when using these SMLM models. The findings of this study can provide valuable insights for policymakers, investors, and researchers who are interested in forecasting inflation rates.

Predicting Commercial Building Energy Consumption Using a Multivariate Multilayered Long-Short Term Memory Time-Series Model

The global demand for energy has been steadily increasing due to population growth, urbanization, and industrialization. Numerous researchers worldwide are striving to create precise forecasting models for predicting energy consumption to manage supply and demand effectively. In this research, a time-series forecasting model based on multivariate multilayered long short-term memory (LSTM) is proposed for forecasting energy consumption and tested using data obtained from commercial buildings in Melbourne, Australia: the Advanced Technologies Center, Advanced Manufacturing and Design Center, and Knox Innovation, Opportunity, and Sustainability Center buildings. This research specifically identifies the best forecasting method for subtropical conditions and evaluates its performance by comparing it with the most commonly used methods at present, including LSTM, bidirectional LSTM, and linear regression. The proposed multivariate, multilayered LSTM model was assessed by comparing mean average error (MAE), root-mean-square error (RMSE), and mean absolute percentage error (MAPE) values with and without labeled time. Results indicate that the proposed model exhibits optimal performance with improved precision and accuracy. Specifically, the proposed LSTM model achieved a decrease in MAE of 30%, RMSE of 25%, and MAPE of 20% compared with the LSTM method. Moreover, it outperformed the bidirectional LSTM method with a reduction in MAE of 10%, RMSE of 20%, and MAPE of 18%. Furthermore, the proposed model surpassed linear regression with a decrease in MAE by 2%, RMSE by 7%, and MAPE by 10%.These findings highlight the significant performance increase achieved by the proposed multivariate multilayered LSTM model in energy consumption forecasting.

Predição da Cotação Real/Bitcoin usando a Rede Neural Long Short Term Memory

Cryptocurrencies have become an important financial asset for com- panies, governments and civilians, being accepted for financial transactions in place of traditional currencies. However, cryptocurrencies, which are attractive for some reasons such as not having government control, have high volatility, making price predictions important in order to have a basis for deciding when to buy and when to sell this type of asset, in order to maximize profits. As pre- diction by people demands a specialist, which is expensive and also subject to biases and errors, several studies have investigated the application of artificial intelligence computational techniques for prediction. This study investigates the application of an LSTM neural network for predicting the price of the Bitcoin cryptocurrency in Reais, analyzing parameters of the neural network, together with the network modeling for the prediction of a single value for a given future time horizon and the prediction of multiple values for multiple future instants of time. The results obtained indicate that the single output model has a good performance for short-term loss, mainly in the prediction for the next day (horizon 1) with a mean absolute percentage error (MAPE) of 2.57% in its best configuration, however this error is not linearly scalable when the prediction horizon is increased. The multiple-output model produced higher MAPE values for short-term predictions (horizons 1, 3 and 5), however the error increase rate is lower than that of the simple horizon model.

Aplikasi Model ARIMA dalam Peramalan Data Harga Emas Dunia Tahun 2010-2022

Gold investment is one of the favorite investments during the Covid-19 pandemic because the price of gold is relatively volatile but shows an increasing trend. Savvy investors investing in gold need to be able to predict future opportunities. Therefore, price estimation is needed to develop a buying and selling strategy to maximize profits. The Autoregressive Integrated Moving Average (ARIMA) model is a suitable method for predicting time series data, so the best ARIMA model will be applied for forecasting world gold prices. The best ARIMA model is selected based on the Akaike Information Criterion (AIC) and Mean Absolute Percentage Error (MAPE) criteria. Monthly world gold price data for 146 periods are applied in this study and will be used to predict gold prices for the following six periods. ARIMA(0,1,1) is the best model obtained from the analysis results, with AIC and MAPE values ​​of 1264.731 and 11.972%, respectively. Forecasting results show that world gold prices will increase for the next periods.

Application of Machine Learning for Prediction and Monitoring of Manganese Concentration in Soil and Surface Water

This study explored the application of machine learning, specifically artificial neural network (ANN), to create prediction models for manganese (Mn) concentration in soil and surface water (SW) on the island province with two open mine pits overflowing to two major rivers that experienced mining disasters. The two ANN models were created to predict Mn concentrations in soil and SW from 12 and 14 input parameters for soil and SW, respectively. These input parameters were extracted from extensive field data collected at the site during sampling program in 2019, 2021, 2022, and initially processed with spatial analysis via geographic information system (GIS). All datasets were then divided for model training and validation, using 85% and 15% ratio, respectively. Performance evaluation of each model with mean absolute percentage error (MAPE) and root mean squared error (RMSE) confirmed the accuracy of both models. The soil Mn model achieved MAPE and RMSE values of 2.01% and 23.98, respectively. The SW Mn model was split into two models based on SW Mn values within the 0–1 mg/L range and >1 mg/L range. The SW Mn model for >1 mg/L performed better with MAPE and RMSE of 4.61% and 0.17, respectively. Feature reduction was also conducted to identify how the models will perform if some input parameters were excluded. Result showed sufficient accuracy can still be obtained with the removal of 4–5 input parameters. This study and these models highlight the benefit of ANN to the scientific community and government units, for predicting Mn concentration, of similar environmental conditions.

Mango quality prediction based on near-infrared spectroscopy using multi-predictor local polynomial regression modeling

Background: pH and total soluble solids (TSS) are important quality parameters of mangoes; they represent the acidity and sweetness of the fruit, respectively. This study predicts the pH and TSS of intact mangoes based on near-infrared (NIR) spectroscopy using multi-predictor local polynomial regression (MLPR) modeling. Herein, the prediction performance of kernel partial least square regression (KPLSR), support vector machine regression (SVMR), and MLPR is compared. Methods: For this purpose, 186 intact mango samples at three different maturity stages are used. Prediction models are built using MLPR, KPLSR, and SVMR based on untreated and treated spectra. The best regression model for predicting pH is MLPR based on Gaussian filter smoothing spectra. Moreover, the TSS value is more accurately predicted using MLPR based on Savitzky–Golay smoothing. Results: The findings reveal that MLPR is highly accurate in estimating the pH and TSS of mangoes, with mean absolute percentage error (MAPE) values less than 10 %. In addition, the MLPR model has the best predictive performance with the lowest Mean Squared error (MSE) and root mean squared error (RMSE) values and the highest R2 value. Conclusions: The use of NIR spectroscopy in combination with multi-predictor local polynomial regression could provide a quick and non-destructive technique for predicting mango quality. Thus, the results of this study help support sustainable production as a sustainable development goal.

Deep learning approach for one-hour ahead forecasting of weather data

ABSTRACT Weather is made up of multiple parameters, including solar radiation (SR), atmospheric pressure (AP), soil temperature (ST), atmospheric temperature (AT), wind speed (WS), relative humidity (RH), and sunshine duration (SD). These factors are also crucial for the renewable energy sector, solar simulation, agriculture, air pollution, water supply and distribution, avalanche warning, forestry, and town and regional planning. A deep learning method based on a neural network with Long Short-Term Memory (LSTM) was employed in this investigation for one-hour-ahead weather data forecasting. The ability of the LSTM model was compared with the Adaptive Neuro-Fuzzy Inference System (ANFIS) with that of the fuzzy c-means (FCM), Autoregressive Integrated Moving Average (ARIMA) model, and the Autoregressive Moving Average (ARMA) model. Mean absolute error (MAE), correlation coefficient (R), root means square error (RMSE), average bias, Nash – Sutcliffe efficiency coefficient (NSE), and mean absolute percentage error (MAPE) were selected as evaluation criteria. Results indicated that the proposed LSTM model presented good enough results compared to other used methods. 7 different types of meteorological data from a total of 4 years (35040 hours) were divided into 25% test data and 75% training data for the models. The best result was obtained for the hourly ST estimation of Adana province using the LSTM method, the MAE, RMSE, R, bias, NSE, and MAPE values were computed as 0.016°C, 0.078°C, 0.9999, −0.00018°C, 0.0805%, and 0.9999, respectively. On the other hand, the worst result was obtained for the hourly SD for Mardin province when ARIMA was used, and the statistical measures were derived as 0.128 hours for MAE, 0.215 hours for RMSE, 0.8851 for R, 0.00091 hours for bias, and 0.7657 for NSE. In this regard, it is demonstrated that the LSTM technique outperformed the other models in terms of all-weather data estimates and delivered highly sensitive outcomes.

COVID-19 PROJECTIONS ON JAVA AND BALI ISLANDS INVOLVING VACCINATION AND TESTING INTERVENTIONS USING VARI-X MODEL

The Indonesian government implemented the policy of increasing vaccination and testing of Covid-19 for travel from or to the Java and Bali Islands to reduce the Covid-19 projected spread in there. As participation in these efforts, this study aims to project the Covid-19 spread measured by the active case rates by involving the intervention of vaccination and testing of Covid-19 in the two islands. Projections are performed using a vector of autoregression integrated with the exogenous variables (VARI-X) model. This model is used because it can simultaneously project the Covid-19 spread in the two islands by involving interventions of vaccination and testing of Covid-19 as exogenous variables. The most suitable model obtained is VARI-X (4, 2, 0). The mean-absolute-percentage error (MAPE) of the model for the Java and Bali Islands is 5.3027% and 3.0301%, respectively. Based on the MAPE value, the model is very accurate for projecting the future Covid-19 spread on the two islands. This accuracy can be seen practically from the Covid-19 spread projection results in the next four days, which are very close to the actual data. This research is expected to help the Indonesian government project the spread of Covid-19 on the Java and Bali Islands.

Deep learning-based prediction for time-dependent chloride penetration in concrete exposed to coastal environment

The application of deep learning methods in civil engineering has gained significant attention, but its usage in studying chloride penetration in concrete is still in its early stages. This research paper focuses on predicting and analyzing chloride profiles using deep learning methods based on measured data from concrete exposed for 600 days in a coastal environment. The study reveals that Bidirectional Long Short-Term Memory (Bi-LSTM) and Convolutional Neural Network (CNN) models exhibit rapid convergence during the training stage, but fail to achieve satisfactory accuracy when predicting chloride profiles. Additionally, the Gate Recurrent Unit (GRU) model proves to be more efficient than the Long Short-Term Memory (LSTM) model, but its prediction accuracy falls short compared to LSTM for further predictions. However, by optimizing the LSTM model through parameters such as the dropout layer, hidden units, iteration times, and initial learning rate, significant improvements are achieved. The mean absolute error (MAE), determinable coefficient (R2), root mean square error (RMSE), and mean absolute percentage error (MAPE) values are reported as 0.0271, 0.9752, 0.0357, and 5.41%, respectively. Furthermore, the study successfully predicts desirable chloride profiles of concrete specimens at 720 days using the optimized LSTM model.

Ensemble framework for daily carbon dioxide emissions forecasting based on the signal decomposition–reconstruction model

The accurate prediction of daily carbon dioxide (CO2) emissions is crucial for grasping the real-time dynamics of CO2 emissions and formulating emission reduction policies. The use of the artificial intelligence model in CO2 emissions prediction has frequently been reported; however, research on the signal decomposition–reconstruction prediction model has rarely been conducted. Daily CO2 emissions are heavily influenced by human activities and show strong non-stationarity, potentially preventing a single artificial intelligence model from yielding satisfactory prediction results. To improve the accuracy of daily CO2 emissions prediction, we propose an ensemble framework based on signal decomposition–reconstruction model for predicting daily CO2 emissions. Our proposed ensemble frameworkis tested on real-world data from 14 regions. The research results show that in predicting daily industrial CO2 emissions, the coefficient of determination (R2) of our proposed model exceeds 0.96, the mean absolute percentage error (MAPE) and root mean square error (RMSE) values are better than those of other models. MAPE is generally within 20% for different forecast lead times. For another kind of CO2 emissions data, our proposed ensemble framework has also demonstrated robust prediction performance for daily ground transport CO2 emissions data, with an R2 exceeding 0.9 in most cases, and a MAPE within 17% for different forecast lead times. This study highlights the efficiency of the proposed model in addressing the issue of daily CO2 emissions prediction. It also provides a method for predicting hourly and annual CO2 emissions.

Pulp Chemistry Variables for Gaussian Process Prediction of Rougher Copper Recovery

Insight about the operation of froth flotation through modelling has been in existence since the early 1930s. Irrespective of the numerous industrial models that have been developed over the years, modelling of the metallurgical outputs of froth flotation often do not involve pulp chemistry variables. As such, this work investigated the influence of pulp chemistry variables (pH, Eh, dissolved oxygen and temperature) on the prediction performance of rougher copper recovery using a Gaussian process regression algorithm. Model performance assessed with linear correlation coefficient (r), root mean square error (RMSE), mean absolute percentage error (MAPE) and scatter index (SI) indicated that pulp chemistry variables are essential in predicting rougher copper recovery, and obtaining r values > 0.98, RMSE values < 0.32, MAPE values < 0.20 and SI values < 0.0034. RNCA feature weights reveal the pulp chemistry relevance in the order dissolved oxygen > pH > Eh > temperature.

Stock Price Prediction of the TMT Industry Based on Machine-Learning

Predicting the TMT stock market enjoys great significance in the financial field and it has been widely researched due to its sophisticated nature and huge potential. Therefore, two machine learning models are trained and tested in this article, i.e., the multi-linear regression model and the random forest model with price information from three stocks, i.e., AAPL, NFLX, and TSLA. Every model used 80% of stock data to train and the rest 20% to test. To evaluate how accurately the models fit the price in reality, the R-squared value and mean absolute percentage error value are used. Firstly, the stationarity had to be checked, which is carried out with the Dicky-Fuller Test. The small p-value showed that the data don’t have a unit root. Then, the data were processed by the two models written by python. According to the analysis, the multi-linear regression and random forest models are relatively accurate. However, with a larger data set or with a relatively small rolling period, the random forest will be able to achieve a better result. However, larger data comes with the price of more computing resources. These results indicate that it is feasible to predict stock market prices with machine learning models and the accuracy could be improved by utilizing better machine learning models.

Evaluation of Field Germination of Soybean Breeding Crops Using Multispectral Data from UAV

The use of multispectral aerial photography data contributes to the study of soybean plants by obtaining objective data. The evaluation of field germination of soybean crops was carried out using multispectral data (MSD). The purpose of this study was to develop ranges of field germination of soybean plants according to multispectral survey data from an unmanned aerial vehicle (UAV) for three years (2020, 2021, and 2022). As part of the ground-based research, the number of plants that sprang up per unit area was calculated and expressed as a percentage of the seeds sown. A DJI Matrice 200 Series v2 unmanned aerial vehicle and a MicaSense Altum multispectral camera were used for multispectral aerial photography. The correlation between ground-based and multispectral data was 0.70–0.75. The ranges of field germination of soybean breeding crops, as well as the vegetation indices (VIs) normalized difference vegetation index (NDVI), normalized difference red edge index (NDRE), and chlorophyll index green (ClGreen) were calculated according to Sturges’ rule. The accuracy of the obtained ranges was estimated using the mean absolute percentage error (MAPE). The MAPE values did not exceed 10% for the ranges of the NDVI and ClGreen vegetation indices, and were no more than 18% for the NDRE index. The final values of the MAPE for the three years did not exceed 10%. The developed software for the automatic evaluation of the germination of soybean crops contributed to the assessment of the germination level of soybean breeding crops using multispectral aerial photography data. The software considers data of the three vegetation indices and calculated ranges, and creates an overview layer to visualize the germination level of the breeding plots. The developed method contributes to the determination of field germination for numerous breeding plots and speeds up the process of breeding new varieties.

Neural-network modeling of solar radiation and temperature variability due to climate change in Ibadan Metropolis

his research focused on studying the variability of solar radiation and temperature under climate change in the Ibadan metropolis. In the study, spatial distribution, temporal variations, annual distribution, estimation and prediction of the solar radiation, minimum and maximum temperature data in the Ibadan Metropolis was collected. A Long Short-Term Memory Neural Network (LSTM-NN) model was developed for the prediction using the time-series data obtained. An ARIMA model was further developed to compare and validate the LSTM-NN model. The performance of the prediction models were determined using the root mean square error (RMSE) and the mean absolute percentage error (MAPE). The RMSE values for the minimum, maximum temperature and solar radiation predictions were 1.543, 1.290, 1.967, and 1.611, 1.309, 2.106 for the LSTM-NN and ARIMA models respectively, while the MAPE values for the minimum, maximum temperature and solar radiation predictions were 3.603, 4.351, 8.859, and 3.840, 4.480, 9.502 for the LSTM-NN and ARIMA models respectively. The LSTM-NN model had a better prediction performance in all categories with lower error when compared with ARIMA. From the prediction, it was observed that there will be a reduction in the maximum temperature, minimum temperature and solar radiation values when compared to obtained data. The observed minimum temperature ranged from 22.9032-23.2032(0C), while the predicted minimum temperature ranged from 19.9260- 19.977(0C) also the observed maximum temperature ranged from 32.87096-33.7064(0C), while the predicted maximum temperature ranged from 29.5159-29.5529(0C), the observed solar radiation ranged from 19.203-19.722 (W/m2 ), while the predicted solar radiation ranged from 14.123-14.115 (W/m2 ). The year with the highest solar radiation which constitutes the useful energy is 2024 with an average value of 14.1395 W/m2

The Plan and Realization Evaluation of Supply Crude Oil Activities at PT Kilang Pertamina International - Refinery Unit VI Balongan

PT Kilang Pertamina International - Refinery Unit VI Balongan is the sixth refinery out of seven refineries owned by PT Pertamina (Persero) that processes crude oil into BBM (Fuel Oil), Non-BBM and Petrochemical products in maintaining energy sustainability in Indonesia. Oil and gas management activities carried out by PT Pertamina must be able to run according to the target. For this reason, in carrying out its business processes, PT Pertamina requires good planning for crude oil. This study focuses on the type of Super Heavy (SH) crude oil including Duri, Cinta, Escalante, Ostra, etc., and Heavy (H) including Banyu Urip, GG Condensate, Minas/SLC, Cabinda, Etame, etc. The purpose of this study is to evaluate the discrepancy between plans and the realization of crude oil orders as well as forecasting for the next 12 periods of crude oil at PT KPI - RU VI Balongan. The results show that several forecasting methods match the demand data pattern, namely the cyclical method, cyclical trend, and multiplicative decomposition (seasonal) methods. In finding the error value forecasting is done by calculating the Mean Absolute Percentage Error (MAPE) method. The selected forecasting method that has the highest accuracy is the cyclical trend forecasting method with a MAPE value of 0.22 for heavy crude oil and 0.13 for super heavy crude oil. The cyclical trend forecasting method has succeeded in reducing the error in the crude supply accuracy level compared to the existing Master Program (MP) crude method.

Correlation and determination of the Metabolizable Energy (ME) of tropical forage with nutrient content for ruminants

The metabolizable energy (ME) of tropical forages measured by in vivo method in ruminants had a high degree of accuracy but requires a long time and is expensive. One method that can be done is the ME estimation model. The objectives of the present study were carried out to investigate the relationship between tropical forage nutrient content and ME for ruminants as well as determine and validate a model for estimating ME of tropical forage based on nutrient content. A total of 26 forage samples consisting of 14 types of grass and 12 legumes were obtained after data pre-processing or data cleaning and data normalization. Forage samples will be grouped into 3, Grass + Legume (G+L=26), grass (R=14), and legume (L=12). The database used is Crude Protein (CP), Extract Ether (EE), Neutral Detergent Fiber (NDF), Acid Detergent Fiber (ADF), and hemicellulose as well as ME with in vivo experiments. The initial stage is preprocessing data. Nutrient content and ME were analyzed using Pearson Correlation and followed by multiple linear regression to determine the ME estimation model. However, validated used the mean absolute deviation (MAD), root means square error (RMSE), and mean absolute percentage error (MAPE). The results showed that there was a significant and highly significantly correlated between nutrient composition and ME in the Grass + Legume, Grass, and Legume groups so it could be used to determine ME. There are 9 regression equations with significance and have high R2 and after being validated with the lowest MAD, RMSE, and MAPE values, three regression equations are obtained with one each for each group Grass + Legume (G+L), Grass (R), and Legumes (L). It is concluded that the regression equation of ME of tropical forage is MER+L = 12.429 – 0.122 ADF for Grass + Legume, EMR = 15.609 – 0.115 NDF for Grass, and EML = 3.726 – 0.186 CP for Legume.