|9 CONCLUSION AND SUMMARY
The existing roads system in Namibia is one of the country's most vital assets. However, one of the overwhelming characteristics of it is the unbalance between the "modern sector" and the structurally underdeveloped areas in the former "homelands" with an undeveloped or even lacking roads system. In the light of the change of socio-economic priorities in the independent Republic of Namibia the restricted resources for the further development of the roads system must be utilised in an optimal manner. Therefore it will be necessary to develop new planning concepts and more appropriate construction as well as maintenance methods in variance to presently used more conventional design and building procedures.
The basis for such new road building methods was the investigation about the status-quo and the structural development of pavements of Namibian roads in chapter 4. Quantifiable analyses like Namibian road statistics served as inventory and introduction. The present situation showed the inequality of the supply and quality of roads between the "modern sector" and other parts of Namibia where the majority of the population lives.
Before any new road building concepts could be realised it was one of the objectives to develop and assemble cost and quality optimised methods for the future construction and maintenance of Namibia's roads system. The basis for these optimised road models has to be the knowledge about Namibia's physical environmental influences on road building which were dealt with in chapters 5 and 6.
These chapters investigated the climatic, geological and soil mechanical influences and gave a survey of all naturally occurring road building materials in the country. It was proved that environmental influences have not a very hard bearing on the properties of road building materials due to the specific features of the Namibian climate. It has, however, a strong influence on the run-off characteristics of catchment areas in Namibia which play an important role in the design of road drainage structures. The consequence of the investigations were more realistic design run-off determination methods.
It can further be concluded that the systematic investigations regarding the occurrences and properties of road building materials will inevitably lead to an improved consciousness under Namibian road designers and builders to continuously deepen this knowledge with the resulting direct saving of considerable public funds.
All occurring road materials in Namibia were dealt with in a summarised form in their geological context. The important road building materials sand and pedocretes were, however, investigated in more detail. It has been, for instance, found that the road building properties of sub-standard calcretes are at least equal to those of the much more expensive crusher-run basecourse. Also the self-stabilising phenomena of pedocrete road building materials like calcretes which have been identified as Namibia's road building material number one was dealt with.
The knowledge about the location of road building materials and their properties served as the basis to the evaluation of pavements for paved and unpaved roads in Namibia. This evaluation lead to the development of cost and quality optimised performance models for construction and maintenance to minimise specific deterioration mechanisms of paved and unpaved roads in chapters 7 to 9.
Because of the general shortage of public funds in Namibia, it has become increasingly necessary to determine and justify construction, rehabilitation and maintenance needs from objective standards. An essential element of any such road management system are the vehicle operating costs ' VOC' in relation to the effect of roughness and pavement type under Namibian conditions. The relations between 'VOC' and road roughness for Namibian circumstances have been founded on 'VOC' studies from other countries like Brazil, India, Kenya, the Caribbean and South Africa. Parameters like fuel consumption, tyre wear, vehicle maintenance and depreciation in relation to different road roughness levels were investigated.
Field tests were undertaken to investigate volumetrical fuel measurements in dependence of road roughness in order to compare this 'VOC' component to that of other studies. A prediction formula for fuel consumption on Namibian roads for different roughnesses was developed. The regression of the evaluated data resulted in a fuel increase of 28,4% for paved and 37,1% for unpaved roads from IRI=0 to IRI=11. The relevant fuel increase for the South African Study was 28,9% for IRI values between 0 and 11 for paved and unpaved roads. The higher slope of the trend curve for unpaved roads against this of paved roads is caused by the higher rolling resistance and increased roughness ( bumpiness) due to tyre temperature (pressure) for unpaved roads. Tests on damp to very wet gravel pavements on the same road sections with similar riding qualities resulted in slightly higher fuel consumptions and a steeper slope of the trend curve against dry unpaved roads due to an increased rolling resistance caused by lower tyre temperatures and suction effects on wet unpaved roads as well as the volumetric change of fuel at different (cooler) temperatures. Thus it was proven that the Namibian values compare reasonably well with those of the South African maintenance design system which will be applicable for Namibia in order to develop cost and quality optimised roads models.
A pavement structure databank formed the basis for optimised performance models in order to simulate the behaviour of different surfaced pavement types under Namibian traffic patterns in chapter 8. It has been shown that such pavement deterioration mechanisms are important due to the fact that the Namibian paved roads are quite unique, even under African conditions, as far as low traffic loads, climatic conditions and unique natural road building materials are concerned. Deterioration mechanisms like loss of surfacing aggregate, bleeding and different forms of crackings were investigated. The objective of this effort was the determination of a quality and cost optimised reseal algorithm and revised materials specifications. It was found that the maintenance of paved roads has been neglected in the last years due to a lack of public funds. The only way to overcome this maintenance backlog lies in the application of a computerised " Pavement Management System".
One result was the road user savings as a function of riding quality of the paved roads system. Road user cost savings are quite minimal for paved roads due to the average high quality of these roads and the extreme low traffic numbers. It was further found that rehabilitation of paved roads cannot be economically justified on ground of vehicle operating costs, but rather on poor riding quality, rutting, skid resistance during wet weather conditions and high routine maintenance costs. It could not be proved that where rehabilitation of surfaced road pavements could not be justified by ' VOC', timing and accident costs would be decisive factors.
Namibia's unpaved roads are named in Africa for their excellent shape and condition, but it could never been proved whether these are effective and cost/quality optimised. The development of such an optimised performance model in chapter 9 was a necessity, because to date, in spite of a treasure of accumulated experience in construction and maintenance of unpaved roads, no consistent and scientifically sound principles for the treatment of unpaved roads existed.
The off-set point for optimised performance models for unpaved roads was the development of limits of acceptance and rejection criteria. Subjective evaluations of specific unpaved roads representing a wide spectrum of different riding qualities were investigated. Contingency table analysis was done to determine the relationship between subjective riding quality evaluation and objective roughness measurements. Regression analysis was used to develop a prediction model for a five point rating in dependence of the user percentile and roughness. The evaluation of the data resulted in a trend that with an average IRI=6,00 the first warnings were expressed, with IRI=7,50 a serious state and with IRI=10,00 the acceptance limit was reached. Other results were that the observed average speeds during the tests didn't reveal any statistical calculable pattern.
A further very decisive part of above acceptability/rejection experiment was the establishment of the main contributing factors to poor road user acceptance of the experimental road sections. The evaluation of the data revealed that a loose layer of sand is the main contributing factor to poor road user acceptance for roads with excellent to good riding qualities while small stones are the main factor for roads with good to average riding qualities and corrugations for roads with average to bad riding qualities. Therefore it could be concluded that corrugation could have a more serious effect on road user acceptance than small stones. Large stones were also judged as contributing factor on bad roads but corrugations still played a greater role. Pot-holes were not regarded as contributing factor to poor road user acceptance on all test sections.
The most important parameter to influence riding quality of unpaved roads is corrugations which appear on nearly all Namibian gravel and earth roads, the road surface being composed of granular material in heaps transverse to the direction of traffic at different wave lengths. The root cause of corrugations would appear to be the loss of soil binder from the wearing course material. The corrugations are further caused by the impact of traffic loads imposed on the road surface. The cure for corrugation is to prevent the drying out of the soil binder with the resulting loss of fines by paving the road or to retain the road surface in position by regular grading. Empirically it was derived that corrugations are a function of cohesion, plasticity, specific gravity and of the percentage of stones in the gravel wearing course as well as the speed and mass of vehicles. The mean corrugation depth recorded for all test sections was 16,9 mm and the mean spacing was 759 mm. It was also found that wave lengths of corrugations are mainly speed dependent while their depths are both speed and material dependent.
It was shown in chapters 7 to 9 that the Namibian unpaved roads system is one of the best in Africa but systematic research to evaluate the advantageous unpaved road sections in order to translate them into adequate material specifications is still in its infancy. To overcome the lack of consistently assembled riding quality data, visual-sensitive inspection data sorted according to acceptance and rejection criteria, were collected and translated into materials characteristics with the objective of a realistic interims materials specification for unpaved roads. Further the experiences learnt from investigations on different deterioration mechanisms like roughness, gravel loss, rut depth, corrugations, dust, pot-holes, cracks, stones, surface drainage and skid resistance during wet weather went into this revised materials specification.
It was found that a wearing course material with all three properties ( PI, GM and maximum size) outside the old specifications has only a 10% chance to be acceptable. The success rate climbs to 37% with two properties outside and to 53% with only one property outside the specification. Less than 7% of the poor sections complied with the full specifications. Further it was found that the most important parameter for quality-optimised wearing courses for unpaved roads are the Grading Modulus 'GM' and the Plasticity Index 'PI'. The investigations included the unique Namibian-style " salt-gravel roads" along the Namibian Atlantic coast line. For these roads it was established to have at least two beneficial, independent material properties like the soluble salt content and/or the 'PI' in order to achieve a satisfactory riding quality. However, the accumulated research effort regarding materials for road building purposes resulted in the criterion that the most important tests for the selection of natural gravels and soils for pavement materials are those for compacted strength at the likely in-service moisture content and not so much those for the determination of grading requirements and Atterberg limits.
The investigations led furthermore to optimised models for grading frequencies and gravel loss replacements as well as the determination of the optimal point to surface an unpaved road based on the investigations for roughness and gravel loss.
The cost/quality optimised point for grading frequencies was established on the ground that for each spent US $ for grading maintenance three US $ in saved user costs should be returned. On this basis the Namibian optimised roads model regarding the optimal point of grading unpaved roads, dependent on 'VOC', roughness and traffic volume was established. The model was based on the accumulated ' VOC' for unpaved roads with different roughnesses from an excellent road with IRI=3,0 to an unacceptable road with IRI=10,5 in dependence of the average daily traffic. The traffic was composed of 85% light and 15% heavy which represents average Namibian traffic conditions.
It was decided to take a road with an IRI=7,50 as an offset point which could be regarded as extreme under Namibian conditions. With average grader maintenance costs of US$ 36/km and with the savings of vehicle operating costs for an unpaved road with an improvement from IRI=7,50 to IRI=5,00 it was derived that for an 'ADT' of 100 a grading of 17 times per year is cost/quality optimised.
The optimal point of regravelling has to be established for each individual material and traffic dependent case. Cost/quality optimised calculations revealed that under average Namibian material conditions and an 'ADT' of 100 (85% light, 15% heavy) 8 to 10 mm gravel material per year have to be replaced. For a gravel layer of average 150 mm thickness this results in a regravelling frequency of more than 15 years. It also has to be borne in mind that this gravel loss is only caused by traffic action and not by environmental factors. Consequently, the optimal point of regravelling could be reached for smaller traffic numbers.
The developed cost/quality optimised system was used to investigate the optimal point of surfacing an unpaved road. This point is a function of the traffic load and the available road building materials which has to be established for each individual case. Average influencing factors were used in order to determine limiting values. The basis of the model was that the total benefits will have to exceed the total costs during the lifetime of a road (20 years). The total costs on a paved road will be equal or less than the total costs on an unpaved road. The developed cost/quality optimised Namibian roads model is based on roughness only. Generally it was found that high traffic volumes will also increase the benefit cost ratio of surfacing unpaved roads.
The model resulted in the following boundary conditions: It would be more advantageous to surface a bad unpaved road to appropriate spoorbaan-level for a traffic load of more than 70 vehicles per day or to another appropriate low-volume level of more than 125 vehicles per day or to a new road level with favourable geometric and drainage properties of more than 190 vehicles per day, while it will be not cost/quality optimised to surface a good unpaved road carrying less than 110 vehicles per day, even for an appropriate low-volume spoorbaan road level. To build a full-scale conventional paved road will only be cost/quality optimised for a bad unpaved road carrying more than 330 vehicles per day or a good gravel road carrying more than 500 vehicles a day. In these evaluations shadow-pricing for technical and economic factors like timing, accident rates, dust and all-weather-trafficability or social factors like labour-intensiveness were not included which could reduce the point of surfacing an unpaved road.
It was found that most of the pavements have been unnecessarily overdesigned in the past with the mentioned resulting proposed new relaxed road material specifications for paved and unpaved roads which results in a more effective resource allocation for roads. The investigations showed clearly that in order to achieve optimal results regarding quality and economy of paved and unpaved roads, it has to be a high priority to make more use of scientific pavement management technologies.
In chapter 10 of this thesis the performance and construction of appropriate low-volume roads were, as a logical succession of the optimised road models, investigated. It was shown that all available resources have to be optimised in order to expand the existing unbalanced roads system between the modern sector and other structurally under-developed regions in Namibia. Another finding was that at least 90% of all existing roads could be classified as low-volume roads and that such roads require unique innovations and design concepts at variance with those of high volume facilities, as, for instance, "Namibia-Adapted Technologies" and innovations such as spoorbaan roads.
Experimental sections for such a new appropriate, low-volume road concept were constructed in spoorbaan technique in order to compare it with the theory and to prove the structural integrity of the system under a given set of parameters. It was found that advantages like labour-intensive, dust-free, no-corrugations, all-weather-trafficability and cost/quality optimised arguments outstrip disadvantages like accident susceptibility due to the one-lane character of the spoorbaan strip road and acceptability problems by the public, although no accidents happened to date on the experimental section. It was found that the structural integrity of this road type behaved as expected under extreme high traffic loads (ADT=300 to 500 with 7,0% heavy vehicles) and high rainfall intensities of the rainy season 1990/91. It was proved that this type of road is particularly suitable for low-cost, low-volume feeder roads in areas deficient of natural good quality road building materials especially in Owamboland and other areas.
The different classes of appropriate low-volume roads were compared with the construction standards as developed by the German " Federal Ministry of Economic Co-operation" for different categories of roads, especially tertiary roads in Africa. It was generally found that the proposed design and construction standards for Namibian appropriate low-volume roads are higher than those of the 'BMZ'. It has, however, to be taken into consideration that the proposed new standards for Namibian low-volume roads must be in balance with conventional road standards in the country. Both standards mirror the dualistic nature of the Namibia-specific form of economy.
It was further found that this concept leads to self-help schemes, based on the mobilisation of local labour which could extend and improve rural road construction, rehabilitation and maintenance. These schemes have been found to be labour-intensive which will be economically advantageous and less import-intensive than conventional road building and maintenance techniques. The problematics between labour and equipment-based technologies were outlined. It was found that no systematic studies to promote labour-based construction have been done to date in Namibia. It was also recommended to initiate labour-based pilot projects as, for instance, the building of spoorbaan roads in the densely populated areas in northern Namibia with very high unemployment rates to prove the technical and socio-economic feasibility of such concepts. A basic appropriateness model for different road building labour-based activities taking into account arguments like economic limits, quality and uniformity assurance as well as socio-economic and psychological factors was developed.