Search results for: MAPE

Authors: Charline David, Alexandre Blondin Massé, Arnaud Zinflou

Abstract:

In 2016, Clements, Hurn, and Li proposed a multiple equation time series approach for the short-term load forecasting, reporting an average mean absolute percentage error (MAPE) of 1.36% on an 11-years dataset for the Queensland region in Australia. We present an adaptation of their model to the electrical power load consumption for the whole Quebec province in Canada. More precisely, we take into account two additional meteorological variables — cloudiness and wind speed — on top of temperature, as well as the use of multiple meteorological measurements taken at different locations on the territory. We also consider other minor improvements. Our final model shows an average MAPE score of 1:79% over an 8-years dataset.

Keywords: short-term load forecasting, special days, time series, multiple equations, parallelization, clustering

Procedia PDF Downloads 69

Authors: Jonas Pedro Fabris, Alberth Almeida Amorim Souza, Maria Emilia Camargo, Suzana Leitão Russo

Abstract:

With the law Nº. 9279, of May 14, 1996, the Brazilian government regulates rights and obligations relating to industrial property considering the economic development of the country as granting patents, trademark registration, registration of industrial designs and other forms of protection copyright. In this study, we show the application of the methodology of Box and Jenkins in the series of deposits of industrial design at the National Institute of Industrial Property for the period from May 1996 to April 2013. First, a graphical analysis of the data was done by observing the behavior of the data and the autocorrelation function. The best model found, based on the analysis of charts and statistical tests suggested by Box and Jenkins methodology, it was possible to determine the model number for the deposit of industrial design, SARIMA (2,1,0)(2,0,0), with an equal to 9.88% MAPE.

Keywords: ARIMA models, autocorrelation, Box and Jenkins Models, industrial design, MAPE, time series

Procedia PDF Downloads 514

Authors: Budi Nurani Ruchjana, Atje Setiawan Abdullah, I. Gede Nyoman Mindra Jaya, Eddy Hermawan

Abstract:

In this paper, we propose the application package to estimate parameters of spatiotemporal model based on the multivariate time series analysis using the R open-source software. We build packages mainly to estimate the parameters of the Generalized Space Time Autoregressive (GSTAR) model. GSTAR is a combination of time series and spatial models that have parameters vary per location. We use the method of Ordinary Least Squares (OLS) and use the Mean Average Percentage Error (MAPE) to fit the model to spatiotemporal real phenomenon. For case study, we use oil production data from volcanic layer at Jatibarang Indonesia or climate data such as rainfall in Indonesia. Software R is very user-friendly and it is making calculation easier, processing the data is accurate and faster. Limitations R script for the estimation of model parameters spatiotemporal GSTAR built is still limited to a stationary time series model. Therefore, the R program under windows can be developed either for theoretical studies and application.

Keywords: GSTAR Model, MAPE, OLS method, oil production, R software

Procedia PDF Downloads 212

Authors: Milan Joshi, Harsh Agrawal, Pallaw Mishra, Sanand Sule

Abstract:

Predicting electricity usage is a crucial aspect of organizing and controlling sustainable energy systems. The task of forecasting electricity load is intricate and requires a lot of effort due to the combined impact of social, economic, technical, environmental, and cultural factors on power consumption in communities. As a result, it is important to create strong models that can handle the significant non-linear and complex nature of the task. The objective of this study is to create and compare three machine learning techniques for predicting electricity load for both the day ahead and intraday, taking into account various factors such as meteorological data and social events including holidays and festivals. The proposed methods include a LightGBM, FBProphet, combination of FBProphet and LightGBM for day ahead and Motifs( Stumpy) based on Mueens algorithm for similarity search for intraday. We utilize these techniques to predict electricity usage during normal days and social events in the Himachal Region. We then assess their performance by measuring the MSE, RMSE, and MAPE values. The outcomes demonstrate that the combination of FBProphet and LightGBM method is the most accurate for day ahead and Motifs for intraday forecasting of electricity usage, surpassing other models in terms of MAPE, RMSE, and MSE. Moreover, the FBProphet - LightGBM approach proves to be highly effective in forecasting electricity load during social events, exhibiting precise day ahead predictions. In summary, our proposed electricity forecasting techniques display excellent performance in predicting electricity usage during normal days and special events in the Himachal Region.

Keywords: feature engineering, FBProphet, LightGBM, MASS, Motifs, MAPE

Procedia PDF Downloads 38

Authors: Wiwat Kittinaraporn, Napat Harnpornchai, Sutja Boonyachut

Abstract:

This paper deals with the application of a novel neural network technique, so-called Support Vector Machine (SVM). The objective of this study is to explore the variable and parameter of forecasting factors in the construction industry to build up forecasting model for construction quantity in Thailand. The scope of the research is to study the non-residential construction quantity in Thailand. There are 44 sets of yearly data available, ranging from 1965 to 2009. The correlation between economic indicators and construction demand with the lag of one year was developed by Apichat Buakla. The selected variables are used to develop SVM models to forecast the non-residential construction quantity in Thailand. The parameters are selected by using ten-fold cross-validation method. The results are indicated in term of Mean Absolute Percentage Error (MAPE). The MAPE value for the non-residential construction quantity predicted by Epsilon-SVR in corporation with Radial Basis Function (RBF) of kernel function type is 5.90. Analysis of the experimental results show that the support vector machine modelling technique can be applied to forecast construction quantity time series which is useful for decision planning and management purpose.

Keywords: forecasting, non-residential, construction, support vector machines

Procedia PDF Downloads 403

Authors: Sizar Abed Mohammed, Zahraa Ghazi Sadeeq

Abstract:

The use of the classic method, least squares (OLS) to estimate the linear regression parameters, when they are available assumptions, and capabilities that have good characteristics, such as impartiality, minimum variance, consistency, and so on. The development of alternative statistical techniques to estimate the parameters, when the data are contaminated with outliers. These are powerful methods (or resistance). In this paper, three of robust methods are studied, which are: Maximum likelihood type estimate M-estimator, Modified Maximum likelihood type estimate MM-estimator and Least Trimmed Squares LTS-estimator, and their results are compared with OLS method. These methods applied to real data taken from Duhok company for manufacturing furniture, the obtained results compared by using the criteria: Mean Squared Error (MSE), Mean Absolute Percentage Error (MAPE) and Mean Sum of Absolute Error (MSAE). Important conclusions that this study came up with are: a number of typical values detected by using four methods in the furniture line and very close to the data. This refers to the fact that close to the normal distribution of standard errors, but typical values in the doors line data, using OLS less than that detected by the powerful ways. This means that the standard errors of the distribution are far from normal departure. Another important conclusion is that the estimated values of the parameters by using the lifeline is very far from the estimated values using powerful methods for line doors, gave LTS- destined better results using standard MSE, and gave the M- estimator better results using standard MAPE. Moreover, we noticed that using standard MSAE, and MM- estimator is better. The programs S-plus (version 8.0, professional 2007), Minitab (version 13.2) and SPSS (version 17) are used to analyze the data.

Keywords: Robest, LTS, M estimate, MSE

Procedia PDF Downloads 206

Authors: Luis C. Parra

Abstract:

The significant wave height prediction is an issue of great interest in the field of coastal activities because of the non-linear behavior of the wave height and its complexity of prediction. This study aims to present a machine learning model to forecast the significant wave height of the oceanographic wave measuring buoys anchored at Mooloolaba of the Queensland Government Data. Modeling was performed by a multilayer perceptron neural network-genetic algorithm (GA-MLP), considering Relu(x) as the activation function of the MLPNN. The GA is in charge of optimized the MLPNN hyperparameters (learning rate, hidden layers, neurons, and activation functions) and wrapper feature selection for the window width size. Results are assessed using Mean Square Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). The GAMLPNN algorithm was performed with a population size of thirty individuals for eight generations for the prediction optimization of 5 steps forward, obtaining a performance evaluation of 0.00104 MSE, 0.03222 RMSE, 0.02338 MAE, and 0.71163% of MAPE. The results of the analysis suggest that the MLPNNGA model is effective in predicting significant wave height in a one-step forecast with distant time windows, presenting 0.00014 MSE, 0.01180 RMSE, 0.00912 MAE, and 0.52500% of MAPE with 0.99940 of correlation factor. The GA-MLP algorithm was compared with the ARIMA forecasting model, presenting better performance criteria in all performance criteria, validating the potential of this algorithm.

Keywords: significant wave height, machine learning optimization, multilayer perceptron neural networks, evolutionary algorithms

Procedia PDF Downloads 66

Authors: Phakamani Xaba, Robert Huberts, Bilainu Oboirien

Abstract:

The present study was aimed at developing & comparing two neural-network-based predictive models to predict shelf-life/product age of South African margarine using free fatty acid (FFA), water droplet size (D3.3), water droplet distribution (e-sigma), moisture content, peroxide value (PV), anisidine valve (AnV) and total oxidation (totox) value as input variables to the model. Brick margarine products which had varying ages ranging from fresh i.e. week 0 to week 47 were sourced. The brick margarine products which had been stored at 10 & 25 °C and were characterized. JMP and MATLAB models to predict shelf-life/ margarine age were developed and their performances were compared. The key performance indicators to evaluate the model performances were correlation coefficient (CC), root mean square error (RMSE), and mean absolute percentage error (MAPE) relative to the actual data. The MATLAB-developed model showed a better performance in all three performance indicators. The correlation coefficient of the MATLAB model was 99.86% versus 99.74% for the JMP model, the RMSE was 0.720 compared to 1.005 and the MAPE was 7.4% compared to 8.571%. The MATLAB model was selected to be the most accurate, and then, the number of hidden neurons/ nodes was optimized to develop a single predictive model. The optimized MATLAB with 10 neurons showed a better performance compared to the models with 1 & 5 hidden neurons. The developed models can be used by margarine manufacturers, food research institutions, researchers etc, to predict shelf-life/ margarine product age, optimize addition of antioxidants, extend shelf-life of products and proactively troubleshoot for problems related to changes which have an impact on shelf-life of margarine without conducting expensive trials.

Keywords: margarine shelf-life, predictive modelling, neural networks, oil oxidation

Procedia PDF Downloads 164

Authors: Óscar Muñoz Garrigós, José Manuel Bernabeu Aubán

Abstract:

Time-sensitive services are the base of the cloud services industry. Keeping low service saturation is essential for controlling response time. All auto-scalable services make use of reactive auto-scaling. However, reactive auto-scaling has few in-depth studies. This presentation shows a model for reactive auto-scaling methodologies with a MAPE-k architecture. Queuing theory can compute different properties of static services but lacks some parameters related to the transition between models. Our model uses queuing theory parameters to relate the transition between models. It associates MAPE-k related times, the sampling frequency, the cooldown period, the number of requests that an instance can handle per unit of time, the number of incoming requests at a time instant, and a function that describes the acceleration in the service's ability to handle more requests. This model is later used as a solution to horizontally auto-scale time-sensitive services composed of microservices, reevaluating the model’s parameters periodically to allocate resources. The solution requires limiting the acceleration of the growth in the number of incoming requests to keep a constrained response time. Business benefits determine such limits. The solution can add a dynamic number of instances and remains valid under different system sizes. The study includes performance recommendations to improve results according to the incoming load shape and business benefits. The exposed methodology is tested in a simulation. The simulator contains a load generator and a service composed of two microservices, where the frontend microservice depends on a backend microservice with a 1:1 request relation ratio. A common request takes 2.3 seconds to be computed by the service and is discarded if it takes more than 7 seconds. Both microservices contain a load balancer that assigns requests to the less loaded instance and preemptively discards requests if they are not finished in time to prevent resource saturation. When load decreases, instances with lower load are kept in the backlog where no more requests are assigned. If the load grows and an instance in the backlog is required, it returns to the running state, but if it finishes the computation of all requests and is no longer required, it is permanently deallocated. A few load patterns are required to represent the worst-case scenario for reactive systems: the following scenarios test response times, resource consumption and business costs. The first scenario is a burst-load scenario. All methodologies will discard requests if the rapidness of the burst is high enough. This scenario focuses on the number of discarded requests and the variance of the response time. The second scenario contains sudden load drops followed by bursts to observe how the methodology behaves when releasing resources that are lately required. The third scenario contains diverse growth accelerations in the number of incoming requests to observe how approaches that add a different number of instances can handle the load with less business cost. The exposed methodology is compared against a multiple threshold CPU methodology allocating/deallocating 10 or 20 instances, outperforming the competitor in all studied metrics.

Keywords: reactive auto-scaling, auto-scaling, microservices, cloud computing

Procedia PDF Downloads 65

Authors: Diteboho Xaba, Kolentino Mpeta, Tlotliso Qejoe

Abstract:

This paper assesses the in-sample forecasting of the South African exchange rates comparing a linear ARIMA model and a SETAR model. The study uses a monthly adjusted data of South African exchange rates with 420 observations. Akaike information criterion (AIC) and the Schwarz information criteria (SIC) are used for model selection. Mean absolute error (MAE), root mean squared error (RMSE) and mean absolute percentage error (MAPE) are error metrics used to evaluate forecast capability of the models. The Diebold –Mariano (DM) test is employed in the study to check forecast accuracy in order to distinguish the forecasting performance between the two models (ARIMA and SETAR). The results indicate that both models perform well when modelling and forecasting the exchange rates, but SETAR seemed to outperform ARIMA.

Keywords: ARIMA, error metrices, model selection, SETAR

Procedia PDF Downloads 216

Authors: Watcharin Sangma, Onsiri Chanmuang, Pitsanu Tongkhow

Abstract:

A forecasting model for steel demand uncertainty in Thailand is proposed. It consists of trend, autocorrelation, and outliers in a hierarchical Bayesian frame work. The proposed model uses a cumulative Weibull distribution function, latent first-order autocorrelation, and binary selection, to account for trend, time-varying autocorrelation, and outliers, respectively. The Gibbs sampling Markov Chain Monte Carlo (MCMC) is used for parameter estimation. The proposed model is applied to steel demand index data in Thailand. The root mean square error (RMSE), mean absolute percentage error (MAPE), and mean absolute error (MAE) criteria are used for model comparison. The study reveals that the proposed model is more appropriate than the exponential smoothing method.

Keywords: forecasting model, steel demand uncertainty, hierarchical Bayesian framework, exponential smoothing method

Procedia PDF Downloads 320

Authors: Kunya Bowornchockchai

Abstract:

The objective of this research is to forecast the monthly exchange rate between Thai baht and the US dollar and to compare two forecasting methods. The methods are Box-Jenkins’ method and Holt’s method. Results show that the Box-Jenkins’ method is the most suitable method for the monthly Exchange Rate between Thai Baht and the US Dollar. The suitable forecasting model is ARIMA (1,1,0)  without constant and the forecasting equation is Y_t = Y_t-1 + 0.3691 (Y_t-1 - Y_t-2) When Y_t  is the time series data at time t, respectively.

Keywords: Box–Jenkins method, Holt’s method, mean absolute percentage error (MAPE), exchange rate

Procedia PDF Downloads 216

Authors: Muhammad Ahnaf Zahin, Yaw Adu-Gyamfi

Abstract:

Bus transit is a crucial component of transportation networks, especially in urban areas. Any intelligent transportation system must have accurate real-time information on bus travel times since it minimizes waiting times for passengers at different stations along a route, improves service reliability, and significantly optimizes travel patterns. Bus agencies must enhance the quality of their information service to serve their passengers better and draw in more travelers since people waiting at bus stops are frequently anxious about when the bus will arrive at their starting point and when it will reach their destination. For solving this issue, different models have been developed for predicting bus travel times recently, but most of them are focused on smaller road networks due to their relatively subpar performance in high-density urban areas on a vast network. This paper develops a deep learning-based architecture using a single-step multi-station forecasting approach to predict average bus travel times for numerous routes, stops, and trips on a large-scale network using heterogeneous bus transit data collected from the GTFS database. Over one week, data was gathered from multiple bus routes in Saint Louis, Missouri. In this study, Gated Recurrent Unit (GRU) neural network was followed to predict the mean vehicle travel times for different hours of the day for multiple stations along multiple routes. Historical time steps and prediction horizon were set up to 5 and 1, respectively, which means that five hours of historical average travel time data were used to predict average travel time for the following hour. The spatial and temporal information and the historical average travel times were captured from the dataset for model input parameters. As adjacency matrices for the spatial input parameters, the station distances and sequence numbers were used, and the time of day (hour) was considered for the temporal inputs. Other inputs, including volatility information such as standard deviation and variance of journey durations, were also included in the model to make it more robust. The model's performance was evaluated based on a metric called mean absolute percentage error (MAPE). The observed prediction errors for various routes, trips, and stations remained consistent throughout the day. The results showed that the developed model could predict travel times more accurately during peak traffic hours, having a MAPE of around 14%, and performed less accurately during the latter part of the day. In the context of a complicated transportation network in high-density urban areas, the model showed its applicability for real-time travel time prediction of public transportation and ensured the high quality of the predictions generated by the model.

Keywords: gated recurrent unit, mean absolute percentage error, single-step forecasting, travel time prediction.

Procedia PDF Downloads 46

Authors: Monika Chuchro

Abstract:

This paper presents a preliminary attempt to apply classification of time series using meta-clusters in order to improve the quality of regression models. In this case, clustering was performed as a method to obtain a subgroups of time series data with normal distribution from inflow into waste water treatment plant data which Composed of several groups differing by mean value. Two simple algorithms: K-mean and EM were chosen as a clustering method. The rand index was used to measure the similarity. After simple meta-clustering, regression model was performed for each subgroups. The final model was a sum of subgroups models. The quality of obtained model was compared with the regression model made using the same explanatory variables but with no clustering of data. Results were compared by determination coefficient (R2), measure of prediction accuracy mean absolute percentage error (MAPE) and comparison on linear chart. Preliminary results allows to foresee the potential of the presented technique.

Keywords: clustering, data analysis, data mining, predictive models

Procedia PDF Downloads 432

Authors: R. Kamarudzaman, A. Kalam, N. A. Mohd Fadzil

Abstract:

Oil Palm Fruit Bunch Fiber (OPEFB) was used as secondary filler in HDPE/clay nanocomposites. The composites were prepared by melt compounding which contains High Density Polyethylene (HDPE), OPEFB fibers, Maleic Anhydride Graft Polyethylene (MAPE) and four different clay loading (3, 5, 7 and 10 PE nanoclay pellets per hundred of HDPE pellets). Four OPEFB sizes (180 µm, 250 µm, 300 µm and 355 µm) were added in the composites to investigate their effects on fracture toughness. Fracture toughness of the composites were determined according to ASTM D5045 and Single Edge Notch Bending (SENB) been employed during the test. The effects of alkali treatment were also investigated in this study. The results indicate that the fracture toughness slightly increased as clay loading increased. The highest value of fracture toughness was 0.47 and 1.06 MPa.m1/2 at 5 phr for both types of clay loading. The presence of filler as reinforcement with the matrix indicates the enhancement of composites compared to those without the filler.

Keywords: oil palm empty fruit bunch, fiber, polyethylene, polymer nanocomposite, impact strength

Procedia PDF Downloads 533

Authors: Azar Khodabakhshi, Elham Bolandnazar

Abstract:

Agriculture as the most important food manufacturing sector is not only the energy consumer, but also is known as energy supplier. Using energy is considered as a helpful parameter for analyzing and evaluating the agricultural sustainability. In this study, the pattern of energy consumption of strawberry greenhouses of Jiroft in Kerman province of Iran was surveyed. The total input energy required in the strawberries production was calculated as 113314.71 MJ /ha. Electricity with 38.34% contribution of the total energy was considered as the most energy consumer in strawberry production. In this study, Neuro Fuzzy networks was used for function modeling in the production of strawberries. Results showed that the best model for predicting the strawberries function had a correlation coefficient, root mean square error (RMSE) and mean absolute percentage error (MAPE) equal to 0.9849, 0.0154 kg/ha and 0.11% respectively. Regards to these results, it can be said that Neuro Fuzzy method can be well predicted and modeled the strawberry crop function.

Keywords: crop yield, energy, neuro-fuzzy method, strawberry

Procedia PDF Downloads 341

Authors: Abdurrasheed Sa'id, Abubakar Isma'il, Waheed Alayande

Abstract:

This study focused on carrying out numerical simulations of Kangimi reservoir sedimentation by reviewing different numerical sediment transport models, and GSTARS3 was selected. The model was developed using the 1977 data. It was calibrated by simulating the 2012 profile and sediment deposition and compared with 2012 hydrographic survey results of NWRI. The model was validated by simulating the 2016 deposition and compared the results with NWRI estimates. Also, the performance of the proposed model was tested using statistical parameters such as MSE (Mean Square Error), MAPE (Mean Average Percentage Error) and R2 (Coefficient of determination) with values of 1.32m, 0.17% and 0.914 respectively which shows strong agreement. After the calibration, validation and performance testing the model was used to simulate the 2032 and 2062 profiles and deposition. The results showed that by 2032 the reservoir will be silted by 25.34MCM or 43.3% of the design capacity and 60.7% of the capacity by the year 2062. A number of sedimentation mitigation measures were recommended.

Keywords: NWRI- national water resources institute, sedimentation, GSTARS3, model

Procedia PDF Downloads 193

Authors: K. Assis, K. P. Chong, A. S. Idris, C. M. Ho

Abstract:

Oil palm or Elaeis guineensis is considered as the golden crop in Malaysia. But oil palm industry in this country is now facing with the most devastating disease called as Ganoderma Basal Stem Rot disease. The objective of this paper is to analyze the economic loss due to this disease. There were three commercial oil palm sites selected for collecting the required data for economic analysis. Yield parameter used to measure the loss was the total weight of fresh fruit bunch in six months. The predictors include disease severity, change in disease severity, number of infected neighbor palms, age of palm, planting generation, topography, and first order interaction variables. The estimation model of yield loss was identified by using backward elimination based regression method. Diagnostic checking was conducted on the residual of the best yield loss model. The value of mean absolute percentage error (MAPE) was used to measure the forecast performance of the model. The best yield loss model was then used to estimate the economic loss by using the current monthly price of fresh fruit bunch at mill gate.

Keywords: ganoderma, oil palm, regression model, yield loss, economic loss

Procedia PDF Downloads 352

Authors: Raed H. Allawi

Abstract:

Formation pressure is the most critical parameter in hydrocarbon exploration and exploitation. Specifically, predicting abnormal pressures (high formation pressures) and subnormal pressure zones can provide valuable information to minimize uncertainty for anticipated drilling challenges and risks. This study aims to interpret and delineate the pore and fracture pressure of the Mishrif reservoir in the southern Iraq Oilfield. The data required to implement this study included acoustic compression wave, gamma-ray, bulk density, and drilling events. Furthermore, supporting these models needs the pore pressure measurement from the Modular Formation Dynamics Tester (MDT). Many measured values of pore pressure were used to validate the accurate model. Using sonic velocity approaches, the mean absolute percentage error (MAPE) was about 4%. The fracture pressure results were consistent with the measurement data, actual drilling report, and events. The model's results will be a guide for successful drilling in future wells in the same oilfield.

Keywords: pore pressure, fracture pressure, overburden pressure, effective stress, drilling events

Procedia PDF Downloads 48

Authors: Ksenija Dumičić, Anita Čeh Časni, Berislav Žmuk

Abstract:

The aim of this paper is to select the most accurate forecasting method for predicting the future values of the unemployment rate in selected European countries. In order to do so, several forecasting techniques adequate for forecasting time series with trend component, were selected, namely: double exponential smoothing (also known as Holt`s method) and Holt-Winters` method which accounts for trend and seasonality. The results of the empirical analysis showed that the optimal model for forecasting unemployment rate in Greece was Holt-Winters` additive method. In the case of Spain, according to MAPE, the optimal model was double exponential smoothing model. Furthermore, for Croatia and Italy the best forecasting model for unemployment rate was Holt-Winters` multiplicative model, whereas in the case of Portugal the best model to forecast unemployment rate was Double exponential smoothing model. Our findings are in line with European Commission unemployment rate estimates.

Keywords: European Union countries, exponential smoothing methods, forecast accuracy unemployment rate

Procedia PDF Downloads 342

Authors: Azzedine Hamza, Chouaib Chakour, Messaoud Ramdani

Abstract:

Energy demand prediction plays a crucial role in achieving next-generation power systems for agricultural greenhouses. As a result, high prediction quality is required for efficient smart grid management and therefore low-cost energy consumption. The aim of this paper is to investigate the effectiveness of a hybrid data-driven model in day-ahead energy demand prediction. The proposed model consists of Discrete Wavelet Transform (DWT), and Adaptive Neuro-Fuzzy Inference System (ANFIS). The DWT is employed to decompose the original signal in a set of subseries and then an ANFIS is used to generate the forecast for each subseries. The proposed hybrid method (DWT-ANFIS) was evaluated using a greenhouse energy demand data for a week and compared with ANFIS. The performances of the different models were evaluated by comparing the corresponding values of Mean Absolute Percentage Error (MAPE). It was demonstrated that discret wavelet transform can improve agricultural greenhouse energy demand modeling.

Keywords: wavelet transform, ANFIS, energy consumption prediction, greenhouse

Procedia PDF Downloads 56

Authors: Dongkwon Han, Sangho Kim, Sunil Kwon

Abstract:

Unconventional gas resources have dramatically changed the future energy landscape. Unlike conventional gas resources, the key challenges in unconventional gas have been the requirement that applies to advanced approaches for production forecasting due to uncertainty and complexity of fluid flow. In this study, artificial neural network (ANN) model which integrates machine learning and data-driven approach was developed to predict productivity in shale gas. The database of 129 wells of Eagle Ford shale basin used for testing and training of the ANN model. The Input data related to hydraulic fracturing, well completion and productivity of shale gas were selected and the output data is a cumulative production. The performance of the ANN using all data sets, clustering and variables importance (VI) models were compared in the mean absolute percentage error (MAPE). ANN model using all data sets, clustering, and VI were obtained as 44.22%, 10.08% (cluster 1), 5.26% (cluster 2), 6.35%(cluster 3), and 32.23% (ANN VI), 23.19% (SVM VI), respectively. The results showed that the pre-trained ANN model provides more accurate results than the ANN model using all data sets.

Keywords: unconventional gas, artificial neural network, machine learning, clustering, variables importance

Procedia PDF Downloads 171

Authors: Watchareeporn Chaimongkol

Abstract:

In this paper, we compare the statistical measures accuracy of composite forecasting model to estimate automobile customer demand in Thailand. A modified simple exponential smoothing and autoregressive integrate moving average (ARIMA) forecasting model is built to estimate customer demand of passenger cars, instead of using information of historical sales data. Our model takes into account special characteristic of the Thai automobile market such as sales promotion, advertising and publicity, petrol price, and interest rate for loan. We evaluate our forecasting model by comparing forecasts with actual data using six accuracy measurements, mean absolute percentage error (MAPE), geometric mean absolute error (GMAE), symmetric mean absolute percentage error (sMAPE), mean absolute scaled error (MASE), median relative absolute error (MdRAE), and geometric mean relative absolute error (GMRAE).

Keywords: composite forecasting, simple exponential smoothing model, autoregressive integrate moving average model selection, accuracy measurements

Procedia PDF Downloads 331

Authors: Alireza Vafaei Sadr, Vida Abedi, Jiang Li, Ramin Zand

Abstract:

Missing data is a common challenge in medical research and can lead to biased or incomplete results. When the data bias leaks into models, it further exacerbates health disparities; biased algorithms can lead to misclassification and reduced resource allocation and monitoring as part of prevention strategies for certain minorities and vulnerable segments of patient populations, which in turn further reduce data footprint from the same population – thus, a vicious cycle. This study compares the performance of six imputation techniques grouped into Linear and Non-Linear models on two different realworld electronic health records (EHRs) datasets, representing 17864 patient records. The mean absolute percentage error (MAPE) and root mean squared error (RMSE) are used as performance metrics, and the results show that the Linear models outperformed the Non-Linear models in terms of both metrics. These results suggest that sometimes Linear models might be an optimal choice for imputation in laboratory variables in terms of imputation efficiency and uncertainty of predicted values.

Keywords: EHR, machine learning, imputation, laboratory variables, algorithmic bias

Procedia PDF Downloads 46

Authors: Mehdi Rostamzadeh

Abstract:

Genetic Algorithms (GAs) are an adaptive heuristic search algorithm premised on the evolutionary ideas of natural selection and genetic. In this study, we apply GAs for fundamental and technical models of exchange rate determination in exchange rate market. In this framework, we estimated absolute and relative purchasing power parity, Mundell-Fleming, sticky and flexible prices (monetary models), equilibrium exchange rate and portfolio balance model as fundamental models and Auto Regressive (AR), Moving Average (MA), Auto-Regressive with Moving Average (ARMA) and Mean Reversion (MR) as technical models for Iranian Rial against European Union’s Euro using monthly data from January 1992 to December 2014. Then, we put these models into the genetic algorithm system for measuring their optimal weight for each model. These optimal weights have been measured according to four criteria i.e. R-Squared (R2), mean square error (MSE), mean absolute percentage error (MAPE) and root mean square error (RMSE).Based on obtained Results, it seems that for explaining of Iranian Rial against EU Euro exchange rate behavior, fundamental models are better than technical models.

Keywords: exchange rate, genetic algorithm, fundamental models, technical models

Procedia PDF Downloads 245

Authors: Marjan Golmaryami, Marzieh Behzadi

Abstract:

Recent developments in stock exchange highlight the need for an efficient and accurate method that helps stockholders make better decision. Since stock markets have lots of fluctuations during the time and different effective parameters, it is difficult to make good decisions. The purpose of this study is to employ artificial neural network (ANN) which can deal with time series data and nonlinear relation among variables to forecast next day stock price. Unlike other evolutionary algorithms which were utilized in stock exchange prediction, we trained our proposed neural network with metaheuristic bat algorithm, with fast and powerful convergence and applied it in stock price prediction for the first time. In order to prove the performance of the proposed method, this research selected a 7 year dataset from Parsian Bank stocks and after imposing data preprocessing, used 3 types of ANN (back propagation-ANN, particle swarm optimization-ANN and bat-ANN) to predict the closed price of stocks. Afterwards, this study engaged MATLAB to simulate 3 types of ANN, with the scoring target of mean absolute percentage error (MAPE). The results may be adapted to other companies stocks too.

Keywords: artificial neural network (ANN), bat algorithm, particle swarm optimization algorithm (PSO), stock exchange

Procedia PDF Downloads 522

Authors: Syazreen N. Shair, Saiful A. Ishak, Aida Y. Yusof, Azizah Murad

Abstract:

In this paper, we forecast age-specific Malaysian mortality rates and life expectancy at births by gender and ethnic groups including Malay, Chinese and Indian. Two mortality forecasting models are adopted the original Lee-Carter model and its recent modified version, the product ratio coherent model. While the first forecasts the mortality rates for each subpopulation independently, the latter accounts for the relationship between sub-populations. The evaluation of both models is performed using the out-of-sample forecast errors which are mean absolute percentage errors (MAPE) for mortality rates and mean forecast errors (MFE) for life expectancy at births. The best model is then used to perform the long-term forecasts up to the year 2030, the year when Malaysia is expected to become an aged nation. Results suggest that in terms of overall accuracy, the product ratio model performs better than the original Lee-Carter model. The association of lower mortality group (Chinese) in the subpopulation model can improve the forecasts of high mortality groups (Malay and Indian).

Keywords: coherent forecasts, life expectancy at births, Lee-Carter model, product-ratio model, mortality rates

Procedia PDF Downloads 188

Authors: Waleed A. Basuliman, Khalid S. AlSaleh, Mohamed Z. Ramadan

Abstract:

Applying the anthropometric dimensions is considered one of the important factors when designing any human-machine system. In this study, the estimation of anthropometric dimensions has been improved by developing artificial neural network that aims to predict the anthropometric measurements of the male in Saudi Arabia. A total of 1427 Saudi males from age 6 to 60 participated in measuring twenty anthropometric dimensions. These anthropometric measurements are important for designing the majority of work and life applications in Saudi Arabia. The data were collected during 8 months from different locations in Riyadh City. Five of these dimensions were used as predictors variables (inputs) of the model, and the remaining fifteen dimensions were set to be the measured variables (outcomes). The hidden layers have been varied during the structuring stage, and the best performance was achieved with the network structure 6-25-15. The results showed that the developed Neural Network model was significantly able to predict the body dimensions for the population of Saudi Arabia. The network mean absolute percentage error (MAPE) and the root mean squared error (RMSE) were found 0.0348 and 3.225 respectively. The accuracy of the developed neural network was evaluated by compare the predicted outcomes with a multiple regression model. The ANN model performed better and resulted excellent correlation coefficients between the predicted and actual dimensions.

Keywords: artificial neural network, anthropometric measurements, backpropagation, real anthropometric database

Procedia PDF Downloads 539

Authors: Martin Doran, Roy Sterritt, George Wilkie

Abstract:

NASA, ESA, and NSSC space agencies have plans to put planetary rovers on Mars in 2020. For these future planetary rovers to succeed, they will heavily depend on sensors to detect obstacles. This will also become of vital importance in the future, if rovers become less dependent on commands received from earth-based control and more dependent on self-configuration and self-decision making. These planetary rovers will face harsh environments and the possibility of hardware failure is high, as seen in missions from the past. In this paper, we focus on using Autonomic principles where self-healing, self-optimization, and self-adaption are explored using the MAPE-K model and expanding this model to encapsulate the attributes such as Awareness, Analysis, and Adjustment (AAA-3). In the experimentation, a Pioneer P3-DX research robot is used to simulate a planetary rover. The sonar sensors on the P3-DX robot are used to simulate the sensors on a planetary rover (even though in reality, sonar sensors cannot operate in a vacuum). Experiments using the P3-DX robot focus on how our software system can be adapted with the loss of sonar sensor functionality. The autonomic manager system is responsible for the decision making on how to make use of remaining ‘enabled’ sonars sensors to compensate for those sonar sensors that are ‘disabled’. The key to this research is that the robot can still detect objects even with reduced sonar sensor capability.

Keywords: autonomic, self-adaption, self-healing, self-optimization

Procedia PDF Downloads 319

Authors: Waleed Basuliman

Abstract:

Anthropometric dimensions are considered one of the important factors when designing human-machine systems. In this study, the estimation of anthropometric dimensions has been improved by using Artificial Neural Network (ANN) model that is able to predict the anthropometric measurements of Saudi males in Riyadh City. A total of 1427 Saudi males aged 6 to 60 years participated in measuring 20 anthropometric dimensions. These anthropometric measurements are considered important for designing the work and life applications in Saudi Arabia. The data were collected during eight months from different locations in Riyadh City. Five of these dimensions were used as predictors variables (inputs) of the model, and the remaining 15 dimensions were set to be the measured variables (Model’s outcomes). The hidden layers varied during the structuring stage, and the best performance was achieved with the network structure 6-25-15. The results showed that the developed Neural Network model was able to estimate the body dimensions of Saudi male population in Riyadh City. The network's mean absolute percentage error (MAPE) and the root mean squared error (RMSE) were found to be 0.0348 and 3.225, respectively. These results were found less, and then better, than the errors found in the literature. Finally, the accuracy of the developed neural network was evaluated by comparing the predicted outcomes with regression model. The ANN model showed higher coefficient of determination (R2) between the predicted and actual dimensions than the regression model.

Keywords: artificial neural network, anthropometric measurements, back-propagation

Procedia PDF Downloads 458