Industrial Engineering and Management

This paper is based on the development of a two staged prediction model using output from experimental machinability studies on machining Chrome Steel in a CNC Machine, so as to predict the MRR (Material Removal Rate) using a two stage prediction process involving application of Artificial Neural Network (ANN) model and then using a Multiple Regression model to refine the results. The input parameters considered for the background study on machinability of chrome steel includes the Cutting speed, Feed rate, Depth of cut and the Coolant proportions.  The effort in the prediction process for the MRR apriori assist the machinist in selection of the best combination the four critical input parameter values that can maximize the MRR. The development study in the current research has deployed a two staged predictions approach. The first stage uses a selection of nine best combinations of the four input parameters. This data set is  then used as input into the ANN Model in three steps of training, testing and predicting the output as the response variable in this case being model for predicting the MRR. The initial data set obtained from the background study was used to train the model using the ANN algorithm. The trained data set obtained was then used to test and subsequently used to predict values of the MRR which indicates the rate of production during the CNC turning operation. The simulated experimental runs were carried out using the four factors at three levels each. Thus a total of 81 trials were run in the simulation platform in the two stage prediction approach. The ANN Model was run on a software platform used for risk assessment and evaluation namely Palisade and later the output of the ANN is used for building an empirical model using the  Multiple Linear Regression(MLR) approach. The regression model was constructed using the Minitab Release 17 platform. The inputs for the MLR model were the parameter setting predicted using ANN. The output summary from the two platforms indicates that coefficient of determination values were 0.9973 (99.73% contribution due to the model). The R-Square (Adjusted) values from the model were to be 99.73% and the R-Square (Predicted) to be 99.71%.  The approach used in this study using the two stage model for predicting machinability of chrome steel as a function of the critical input parameters, clearly demonstrate the level of accuracies that can be obtained using intelligent algorithms. The study has significance in evolving the recommendation for setting the parameters during the CNC turning operations on any materials used during research investigations. This study can be further extended by incorporating a multistage refinement using various other intelligent algorithms in combination, so as to further improve the accuracies of predictions.

The role of Intellectual Property and its importance as an asset for business development, nation building has been quite evident in the recent times. The impact of educational and research institutions has been recognized globally as a key source for development of intellectual property and for new technology development. This paper explores how intellectual property adds value to a technical education institution. The other aspect explored is the means by which IP is valuated and the importance of IP valuation.

Emergence of India as economic power can be attributed to the sustained efforts in the field of capacity building, technological focus and innovation. Though these initiatives are being implemented in the best possible manner, they are unable to deliver the same impact as seen in other developing counties in the Asia Pacific region which accounted similar conditions as India a few decades earlier. This paper studies the present condition of patenting in India in comparison to other countries in Asia Pacific region. One of the possible reasons for this can be the prioritization of Intellectual property as a facilitating tool in Innovation and technological advances. Intellectual Property Rights has established itself as a key enabler in promoting, protecting and nurturing the technological and research initiatives which are mandatory for the economic growth.

This paper is based on the development of a two staged prediction model using output from experimental machinability studies on machining Chrome Steel in a CNC Machine, so as to predict the MRR (Material Removal Rate) using a two stage prediction process involving application of Artificial Neural Network (ANN) model and then using a Multiple Regression model to refine the results. The input parameters considered for the background study on machinability of chrome steel includes the Cutting speed, Feed rate, Depth of cut and the Coolant proportions.  The effort in the prediction process for the MRR apriori assist the machinist in selection of the best combination the four critical input parameter values that can maximize the MRR. The development study in the current research has deployed a two staged predictions approach. The first stage uses a selection of nine best combinations of the four input parameters. This data set is  then used as input into the ANN Model in three steps of training, testing and predicting the output as the response variable in this case being model for predicting the MRR. The initial data set obtained from the background study was used to train the model using the ANN algorithm. The trained data set obtained was then used to test and subsequently used to predict values of the MRR which indicates the rate of production during the CNC turning operation. The simulated experimental runs were carried out using the four factors at three levels each. Thus a total of 81 trials were run in the simulation platform in the two stage prediction approach. The ANN Model was run on a software platform used for risk assessment and evaluation namely Palisade and later the output of the ANN is used for building an empirical model using the  Multiple Linear Regression(MLR) approach. The regression model was constructed using the Minitab Release 17 platform. The inputs for the MLR model were the parameter setting predicted using ANN. The output summary from the two platforms indicates that coefficient of determination values were 0.9973 (99.73% contribution due to the model). The R-Square (Adjusted) values from the model were to be 99.73% and the R-Square (Predicted) to be 99.71%.  The approach used in this study using the two stage model for predicting machinability of chrome steel as a function of the critical input parameters, clearly demonstrate the level of accuracies that can be obtained using intelligent algorithms. The study has significance in evolving the recommendation for setting the parameters during the CNC turning operations on any materials used during research investigations. This study can be further extended by incorporating a multistage refinement using various other intelligent algorithms in combination, so as to further improve the accuracies of predictions.