A comparative summary of four methods from the literature review was carried out. It covers:

This review covers several regions of the world: North America, North Africa, Central Africa and Europe. No similarly studied sites were found in West Africa. The comparison covers data collection methods, data processing and the production of thematic maps.

Based on their nature, the data originate from 4 sources:

- Base maps [1] [4] .

- Aerial images [7] .

- Geotechnical, geological and environmental study reports [1] [4] [7] .

- Databases [3] [4] .

Recourse to any one of these data sources depends on its availability in the country’s historical archives. These data are generally available in developed countries: Quebec [7] , Algeria [1] and Belgium [4] . However, this basic data is lacking in developing countries such as Congo [3] .

Data processing leads to the determination of the parameters to be mapped: in the four cases studied, [1] and [3] carried out the geotechnical classification of soils according to several international systems (LCPC, GTR or AASHTO). In 2006, [4] . Digitized and modeled soil stratigraphy using borehole data. [7] also determined soil stratigraphy.

In other words, there are two key parameters for understanding soil: soil class and stratigraphy.

Except for [7] , who drew up a manual map, all three authors drew up the map using ArcGIS GIS and cartography software.

This comparative study shows that there is convergence in the methods used to collect, process and establish the maps. The use of particular data depends on the data sources and the means made available for data collection. Once the data has been collected, two parameters can be used to identify the soil. These are soil class and stratigraphy. These parameters, together with the geodetic coordinates, are used to draw up the maps.

In Benin, there is no geotechnical mapping that could be updated, and the existing geological map has not been updated. On the other hand, geotechnical prospecting data is available from a number of players: geotechnical laboratories, contractors and project owners. This data is acquired repeatedly for each infrastructure construction project. The methodological approach is therefore as follows:

The materials are:

The selection of a platform is guided by the need for it to be open and accessible to any operator approved by the national geotechnical authority to access and record their borehole data in real time. The platform design consists of a database deployed on a secure cloud server and processing and mapping applications.

Table 1 below describes the entities in the conceptual data model.

The diagram of the conceptual model of the database is shown in the following figures (see Figure 2 ).

The ROAD MAT platform is a user-friendly application for recording data in the database. It has an authentication interface, a welcome interface and interfaces for recording detailed data. Figure 3 shows a model interface for the ROAD MAT platform. It can be used to sort data and export results to different file formats (Word, Excel, KMZ, Shapefile).

To process the data, the geotechnical classes and the California Bearing Ratio (CBR) of the materials taken from the soil quarries were determined. Soil classification and bearing capacity are used to assess the geomechanical quality of granular materials that can be used for pavement layers. The classification method used is that of the Centre Expérimental de Recherches et d’Etudes des Bâtiments et des Travaux Publics (CEBTP) [8] . This classification was used to classify laterites from Côte d’Ivoire and the Democratic Republic of Congo [9] . It has the advantage of dividing laterites into three classes according to their decreasing geomechanical qualities (G1, G2, G3). The materials sampled in the borehole are assigned to each class as follows:

Considering:

The percentage passing the 0.08 mm sieve obtained by particle size analysis tests (standard NF P 94-056), noted (f);

- The plasticity index obtained by the Atterberg limit test in accordance with standard NF P 94-051, noted IP.

- The percentage passing the 0.08 mm sieve obtained by particle size analysis tests (standard NF P 94-056), noted (f);

- The plasticity index obtained by the Atterberg limit test in accordance with standard NF P 94-051, noted IP.

The class of the material: G1 - G2 - G3 (see Figure 4 ); a material is of class

o G1 if f ≤ 15% for IP < 16 and f*IP < 250 when IP > 16

o G2 if 15% < f < 25%, and 250 < f*IP < 600 and f*IP < 250 when IP > 16

o G3 if 25% < f < 35%, and 600 < f*IP < 1000 when IP > 28

Type G1 is the best laterite gravel with a CBR of over 30, an OPM dry density of over 2.1 and a water content of between 5 and 8; Group G2 is made up of lower quality laterites with a CBR of between 15 and 30, an OPM dry density of between 2 and 2.1 and an optimum compaction water content of between 7 and 10, while Group G3 is made up of laterites with a CBR of less than 15, an OPM dry density of between 1.9 and 2.2 and an optimum compaction water content of between 8 and 12 (see Figure 5 ).

In addition, three classes of CBR were defined by test in accordance with standard NF P 94-093 as follows:

- 5 < CBR < 30, materials accepted as subgrade; noted P1.

- 30 < CBR > 80, materials accepted as sub-base layer; noted P2.

- CBR > 80 materials accepted as base course, noted P3.

In order to validate the methodology, the data recorded relates to a selection of 714 km of linear roads in six departments, surveyed between 2014 and 2023 by the three main laboratories. However, Table 2 of results for the Mono department is presented below. It covers 16 holes drilled in laterite quarries.

From the table it can be seen that of the 16 boreholes, 4 belong to the G1 class, 11 to the G2 class and 1 to the G3 class, while 15 belong to the CBR P2 class and 1 to the CBR P1 class. We can deduce that the materials in this area of the country are of average geomechanical quality G2 and also of average CBR P2. This result provides geotechnical reference data for pavement design.

It used the ArcGIS 10.8 software to design and produce the granular soil class and CBR class maps. An extract of the results is shown in Figure 5 .

Analysis of the maps gives an idea of the geographical dispersion of the boreholes associated with their geomechanical quality. Class G2 and P2 are dominant on both maps. It should also be noted that the geotechnical surveys were carried out in the same area (in this case, the commune of Houeyogbe). Surveys should be carried out in areas not covered by a materials inventory campaign to supplement the data provided by the surveys carried out for the roadworks.