There are many variables involved in the real life problem so it is difficult to choose an efficient model out of all possible models relating to analytical factors. Interaction terms affecting the model also need to be addressed because of its vital role in the actual dataset. The current study focused on efficient model selection for collector efficiency of solar dryer. For this purpose, collector efficiency of solar dryer was used as a dependent variable with time, inlet temperature, collector average temperature and solar radiation as independent variables. Hybrid of the least absolute shrinkage and selection operator (LASSO) and robust regression were proposed for the identification of efficient model selection. The comparison was made with the ordinary least square (OLS) after performing a multicollinearity and coefficient test and with a ridge regression analysis. The final selected model was obtained using eight selection criteria (8SC). To forecast the efficient model, the mean absolute percentage error (MAPE) was used. As compared to other methods, the proposed method provides a more efficient model with minimum MAPE.
Solar drier is considered to be an important product used in the internet of things (IoT). It is used to dry different kinds of products used in agriculture or aquaculture. There are many factors that have different effects on the drying of items in the solar drier. The current study focused on the removal of the moisture ratio in the drying process for seaweed using solar drier. For this purpose, a dataset containing 1924 observations was used to study the effect of six different independent variables on the dependent variable. Moisture ratio removal (%) was considered to be dependent variable with ambient temperature, chamber temperature, collector temperature, chamber relative humidity, ambient relative humidity and solar radiation as independent variables. All possible models were used in the analysis till fifth order interaction terms. Hybrid model of LASSO with bisquare M was proposed for efficient selection of the model. The procedure based on four phases was used for efficient model selection and a comparison was made with other existing sparse and robust regression techniques. The result indicates that the proposed technique is better than other existing techniques in terms of mean squared error (MSE) and mean absolute percentage error (MAPE).
Application of the Internet of things (IoT) for data collection in solar drying can be very efficient in collecting big data of drying parameters. There are many variables involved so it is hard to find a model to predict the moisture content of the food product during drying. In model building, interaction terms should be incorporated because they also contribute to the model. Eight selection criteria (8SC) is a very useful method in model building. This study applied ordinary least squares (OLS) regression and ridge regression with 8SC in model building to predict the moisture content of drying fish. A total of eighty models were considered in this study. One best model was chosen each from OLS regression and ridge regression. M78.7.3 with a total of eleven independent variables was the best OLS model after conducting multicollinearity and coefficient test. Next, the best ridge model M56.0.0 was obtained after the coefficient test. The mean absolute percentage error (MAPE) was used to measure the accuracy of the prediction model. For OLS model M78.7.3, the MAPE value was 15.7342. The MAPE value for ridge model M56.0.0 was 17.4054. From the MAPE value, OLS model M78.7.3 provided a better estimation than the ridge model M56.0.0. However, OLS model M78.7.3 violated the normality assumptions of residuals. This is highly caused by the outlier problem. So, due to non- normality of the residuals and presence of outliers in the dataset, ridge regression is preferred for the best forecast model.
The economic production quantity (EPQ) model for delayed deteriorating items considering two-phase production periods, exponential demand rate and linearly increasing function of time holding cost is proposed to solve a production problem similar to the one caused by the Covid-19 pandemic. Without shortages, the necessary and sufficient conditions for optimality of this model are characterized through a theorem and lemmas while a solution methodology based on differential calculus is adopted. This paper determines the best replenishment cycle length corresponding to the optimal total variable cost and production quantity of imperfect production industry. To illustrate this model, a numerical experiment is conducted. The results demonstrate that a higher carrying charge decreases the production quantity and a longer demanding period decreases the total variable cost of an industry with a distracted production period. Finally, managerial insights are discussed using sensitivity analysis and future research directions are exposed.
The linear regression is critical for data modelling, especially for scientists. Nevertheless, with the plenty of high-dimensional data, there are data with more explanatory variables than the number of observations. In such circumstances, traditional approaches fail. This paper proposes a modified sparse regression model that solves the problem of heterogeneity using seaweed big data as a use case. The modified heterogeneity models for ridge, LASSO and Elastic net were used to model the data. Robust estimations M Bi-Square, M Hampel, M Huber, MM and S were used. Based on the results, the hybrid model of sparse regression for before, after, and modified heterogeneity robust regression with the 45 high ranking variables and a 2-sigma limit can be used efficiently and effectively to reduce the outliers. The obtained results confirm that the hybrid model of the modified sparse LASSO with the M Bi-Square estimator for the 45 high ranking parameters performed better compared with other existing methods.