چکیده:
Estimation (Forecasting) of industrial production costs is one of the most
important factor affecting decisions in the highly competitive markets. Thus,
accuracy of the estimation is highly desirable. Hibrid Regression Neural Network
is an approach proposed in this paper to obtain better fitness in comparison
with Regression Analysis and the Neural Network methods. Comparing the estimated
results from Regression Analysis and Neural Networks with the Hybrid
Neural-Regression method has indicated the superiority of the latter method.
خلاصه ماشینی:
"Keywords: Regression Analysis, Neural Networks, Hybrids Neural–Regression Method, Comparison, Estimation I- Introduction Cost estimation is performed regularly throughout the life cycle of many products.
Using the Boothroyed and Dewhurst (1991) approach as the source of data, Shtub and Zimmerman (1993) developed a neural network model that estimates the cost of assembly by different assembly systems.
It is completed after presentation of each input – output vector pair from the training set to the GRNN input layer only once, that is, both the centers of the redial basis functions of the pattern unit and the weights in connections of the pattern units and the processing units in the summation layer are assigned simultaneously.
The weights in connection between each pattern unit and the individual summation units are directly assigned with values identical to the elements of the output vector corresponding in the training set to the input vector which formed the center of the redial basis function of that particular pattern unit.
In designed GRNN, the output of neural network (NN) & regression analysis (RA) -that are available in Shtub - are considered as input vectors and actual cost is defined as target of GRNN.
] Figure 4: The Proposed Hybrid Neural-Regression Approach 4- Comparative Analysis The results of Cost estimation is evaluated based on three methods, that of the regression analysis and neural network results are used from Shtub (1999, p.
3- Shtub, Versano (1999), "Estimating the Cost of Steel Pipe Bending, a Comparison between Neural Networks and Regression Analysis", International Journal of Production Economic, Vol. 62, PP."