Page 22 - Microsoft Word - 20120528 Unesco study
P. 22

2.3.6  Costs
                  This study will focus on direct costs; therefore the following cost specifications are required:
                      -   Construction costs for different road constructions.
                      -   Construction costs of the road constructions on the basis of the use of the PowerCem
                          technology.
                      -   Damage costs.
                      -   Maintenance costs.

                  These costs will be calculated in collaboration with PowerCem Mexico using reference projects
                  in the state Tabasco.

                  The indirect costs are expected to be high and will have an enormous influence on the macro-
                  economic effects. When a road is damaged such that it has to be closed, the impact on the
                  population or industry can be high. Due to the limited timeframe of this study, these costs are
                  not included in the model. The available information, though, will be used to give an
                  impression of the expected indirect costs and will be used for recommendations for further
                  research.


                  2.4  Verification
                  The input data of a model must be verified. The input data is the foundation for the output
                  that will come from that model. The input data must be verified by several parties. Also, the
                  sensitivity of a variation in the input data on the total model must be verified. Certain input
                  data has a major influence on the final results, and this input data must be verified with careful
                  attention. The input data is verified by SCT, PowerCem Mexico, PowerCem Technologies and
                  the contractors.


                  2.5  Validation
                  After the input data is verified and the output is known, relations will be laid between the
                  input variables and the output. The validation is split into three procedures:
                      1.  The input variables, must lead to a model output that will conform to the actual
                         results.
                      2.  If other input variables are used for another region, the output of the model must also
                         lead to output that is correct for this other region.
                      3.  By changing the input with high variations, the output of the model must also indicate
                         that this results in high variations in the output.






















                                                                                                     21
   17   18   19   20   21   22   23   24   25   26   27