الفهرس 
chapter 1Only 14 pages are availabe for public view
 |  | from | 153 |  |  |
| | | from | 153 | | |
Abstract
Developing a new technique for pre-tender cost estimate of agriculture
subsurface drainage projects is the aim of this study. Through developing a
model that is able to predict cost of subsurface projects at early stages of
project design before preparing the detailed bill of quantity as an
alternative technique instead of traditional one, which will be a helpful tool
for drainage consulting firm in Egypt. Also it is considered as judgment tool
to determine priority of implementing projects according to available
general budget for projects which have the same conditions. Several
techniques were adopted carefully to identify the factors that affect cost of
subsurface drainage projects at pre-tender stage through reviewing literature
studies, bill of quantities and expert’s interviews. Forty one factors were
collected and grouped in six groups. A questionnaire survey was agreed out
through sixty eight qualified subsurface engineers and contractors in Egypt
to get the most important factors. Based on the result of this survey, twelve
factors that have the greatest effect were identified. These factors are:
National rules and regulations, feedback information from previous
projects, leveling project area on new updated maps and accuracy of maps
information, type of donor finance, inflation rate, area of land to be
drainaged, quantity of laterals used, quantity of plastic collector drains used,
quantity of reinforced concrete collector drains, quantity of pitching with
stones and mortar, number of precast reinforced concrete manholes, and
another requirement cost for construction site. These most important factors
were used for developing models by using regression analysis and artificial
neural network. Application SPSS version 19.0.0 and Neural Power
Professional Version 2.50 were used for developing the desired models.
These models used the twelve input parameters to predict cost of subsurface
drainage (output). The required information collected from 61 real