Civil Engineering April 2022 | Vol 30 No 3

Civil Engineering April 2022 27 of data fields to other data fields. This is called a relational database. Relating different test results and readings to a spe- cific test pit, layer, or site will significantly improve the ability of the DGE to improve decision making through a more efficient comparison of these values. Most of the available software programs or packages had some criteria for the DGE, but none of them were a complete fit (see Figure 1). An essential requirement for a DGE in South Africa is implementing SANS3001 in the test methods. The comparison showed that none of the available software (GEOSYSTEM, Gaea Software, gINT, Kentucky Transportation Basemap Service and SVSoils) allows the use of SANS3001 in their calculations. The COLTO classifi- cation was also not available in any of the programs, most of which used codes spe- cific to their country of origin or standard codes used by several countries, such as ASTM or Eurocodes. In addition, some software is only compatible with other databases such as CivilTech and Novotech (which usually work with gINT). DESIGNING THE DATABASE Microsoft Access was used to create a database to store all the data and information of the important tests. This was selected based on the ease of use and general accessibility of Microsoft Office tools to professionals expected to make use of this software. Microsoft Access is designed to link effectively with other Microsoft Office tools. Microsoft Office also has the advantage of Visual Basics for Applications (VBA) which allows automa- tion and improved customisation. The Microsoft Access database was made with tables, fields, records, queries and forms. For each of the required tests, a table was made with fields for each of the inputs. Each of the tables was organised using letters at the beginning of each name. The letters were used to sort the tables in alphabetical order for organisational purposes. These tables and their relationships are the foundation of the database. Each of these tables with the fields and records al- lows querying of data for analysis. This was significant to the DGE that requires certain information from several tables for com- parison when a decision needs to be made using specific data from multiple tests. The data collected was categorised into specific tests. The first step was to design the tables, which were made for each laboratory test or any category of geotechnical investigations that needed to be used by the DGE. The first table stores the important information about the sample and became the central table from which most tables were directly or indirectly related (see Figure 2). This was used for filtering and querying purposes. Fields were then created in each table to allow each record to produce the correct information. The most important fields are the primary keys, which distinguish the test, and the field that related the test to a specific sample, the foreign key. The rest of the fields corresponded with the required inputs of each test/‌category. Relationships are vital to preventing data loss and are required to make the database a relational database manage- ment system (RDMS). Once the tables with their fields had been created, they needed to be linked through relationships. Most tables were related to the sample information table. The relationships be- tween the tables and the sample data table allowed each sample to have more than one result. These relationships are called one-to-many relationships. Some tests required two or more tables to allow for tests that produce multiple values within the same procedure. This was done by making parent tables that provide the key information for the tests with a child table linked to the parent table to allow multiple records to relate to the test procedure indicated in the parent table. These relationships between the parent test table and the second table were designed as one-to-many relationships as well. After the relationships had been cre- ated, the core database was completed. The relationships between all the tables within the database are shown in Figure 2. In addition, all the main tables are linked to the sample data table to allow any sample and any related informa- tion to be queried. CONCLUSION With the DGE in place, data can now be stored and transformed for decision-making. The literature review about available information systems provided much-needed information about the current systems used and the availability of these systems for South African engineers. During this study it became clear that an information system should be put into place to assist with the ef- fective management of data that has spatial information. The system should transform the data into the required outputs of the tests (especially SANS3001 standard tests) and give important engineering insight that assists decision-making. Unfortunately, the available systems do not provide much assistance to deci- sion-makers, especially in a South African context. For this reason, the DGE was created for engineers to store, manage and compare relative geotechnical engineering information. The next stage for the DGE will be building a network of laboratories and engineers to contribute data to the database and share the information with all professionals in South Africa.  REFERENCES De, A. 2015. Geotechnical site characterization. s.l.:Momentum Press. South African National Roads Agency. 2013. South African Pavement Engineering Manual Chapter 1 Introduction. 1.0 ed. s.l.:SANRAL Ltd. The next stage for the DGE will be building a network of laboratories and engineers to contribute data to the database and share the information with all professionals in South Africa. Microsoft Access was used to create a database to store all the data and information of the important tests. This was selected based on the ease of use and general accessibility of Microsoft Office tools to professionals expected to make use of this software. Microsoft Access is designed to link effectively with other Microsoft Office tools. Microsoft Office also has the advantage of Visual Basics for Applications (VBA) which allows automation and improved customisation.