**4.2 Application of neural networks to determine the rate of internal logistics of an industrial company. Case studies**

Validation errors of the neural network are shown in **Figure 10**.

#### **4.3 Internal logistic indexes of the studied companies**

The ANN enabled in MATLAB with data values of the 13 Internal Logistic Properties from the 10 companies was processed. The values of the indexes of Internal Logistics as well as their possible are given in **Table 10**.

#### **5. Conclusions**

In this chapter three approaches and their expressions to assess the internal logistics of a company are established. The first method was based on dividing the internal logistics in 13 properties, having each property 10 indicators that were evaluated between 1 and 5 points. This leads to the maximum value of Internal Logistics Index (ILI) for each company can reach up to 100 points according to the weight stablished for each indicator.

The second approach was based on the fuzzy logic and the third one was based on neural networks.

When assessing the Internal Logistics Index using the Excel tab developed or by the method of Artificial Neural Networks, very similar values consistent with the reality of the companies studied were obtained, demonstrating the validity of both methods.

The methodological approach developed for the definition of the three models contains all the steps and procedures, allowing replication of the research, which is as important as the application of the developed models at the companies.

When assessing this parameter, using the Excel tab or developed by the method of fuzzy logic, similar values were obtained in line with the reality of the company analyzed, indicating the rationality of both methods.

The approach through the Fuzzy logic allows assess the rate of internal logistics for any position of the input variables, which can obtain a value between 0 and 10, depending on the appreciation of the user of this procedure. In the case of the company studied, specialists gave more emphasis to groups A and C.
