**1. Introduction**

The quality of an asphalt road surface depends on many factors, including: The type of materials used (aggregates, asphalt, and others), climatic conditions, tra ffic load, the drainage system, and the method of construction [1,2]. Important factors directly determining the service life of the road surface are the features of the designed asphalt mix, such as: fatigue limit, resistance to rutting, or resistance to water and frost [3–5]. Other parameters, such as the density of the mineral-asphalt mixture or the air void content (porosity), are indirect. It is generally assumed that in case of dense-graded asphalt mix the content of air voids should be high enough to avoid rutting and low enough at the same time to limit water and dirt penetration into the surface structure [6]. In addition, the surface texture is related to the content of air voids in the wearing layer, which has a direct impact on the safety of road users [7]. Hence, the need for appropriate compaction of particular pavement layers has been known since the beginning of asphalt road construction technology. The main purpose of asphalt mix compaction is to achieve appropriate density to meet the requirements for physical and mechanical properties and to ensure a tight surface which will be exploited for the maximum possible life cycle [8]. It is a di fficult process involving the selection of appropriate compaction parameters in order to achieve the assumed compaction index (CI) of the layer.

Under laboratory conditions, samples of asphalt mix can be compacted in various ways, depending on the purpose of the sample (type of tests), the applicable regulations, as well as the type

of technology used [9]. The most commonly used devices are gyratory compactor, Marshall compactor, asphalt roller compactor, and a vibrating compactor [10]. These methods di ffer in many parameters, such as the pressure force, the way the force is transmitted, compaction time, and also the final shape of the obtained sample and the particles orientation inside [11]. In many European countries the basic method is Marshall compaction method which involves dropping the compacting hammer on a sample of asphalt mix in cylindrical form a certain number of times, which depends on the further use of the sample. Another way is to use a gyratory compactor, where the sample is compacted to the required density or a certain number of rotations, while being simultaneously compacted and sheared [12]. The method is essential in the case of projecting asphalt concrete mixes for very thin layers (BBTM). Moreover, A. Woszuk stated that in order to determine the operating temperatures of Warm Mix asphalt the gyratory compaction method is recommended [10]. According to the European standards [13], the sample placed in a cylindrical form with a diameter of 100 mm or 150 mm is subjected to a constant compressive stress of 600 kPa during the rotational movement, which is transmitted through the piston to parallel sample bases. During the compaction, the sample is rotated at a speed of 30 rpm and the angle of inclination is 1.25◦ (typical parameters in the Superpave system [14]). Such process better reflects the compaction in real conditions than Marshall method as well it is possible to determine significant coe fficients concerning susceptibility to compaction of the mix: compaction coe fficient K, mixture stability index MSI, and mixture resistance index MRI [15,16]. Georgiou et al. found that in the case of compacting in the gyratory device, the samples were segregated, whereas in the case of cores they had a low level of segregation [17]. Additionally, in order to obtain the internal structure of samples comparable to roller compaction, the compaction angle parameter should be appropriately selected, slightly higher than that specified by technical regulations. Taking into account the resistance to rutting, Tapkin and Keskin demonstrated that the samples made in the gyratory press were less susceptible to permanent deformations than those compacted using the Marshall method [18]. Lee et al. found that in case of crumb rubber asphalt mixes air void content was significantly dependent on the number of gyratory rotations and directly related to this rutting increased along with air voids [19]. In reference to the interlayer bonding, Jaskuła showed that the higher the compaction index, the higher was maximum shear strength between the layers [20]. The shearing and mechanical behavior is one of the major issues in many other fields like food industry, paints and coating industry, and also energy industry. In these areas hybrid gels featuring interpenetrating covalent can be applied, which exhibit both high toughness and recoverability of strain-induced network damage [21].

In the rutting test, however, samples made in a roller compactor are used [22]. Airey and Collop indicate that slab-compacted specimens are more closely correlated with cores sampled from the road surface than gyratory and vibratory-compacted specimens [11]. There are also self-compacting mixes, e.g., mastic asphalt, often used in bridge structures [23] or SCC cement concrete [24].

Under real construction conditions, the target compaction is carried out by road rollers of di fferent sizes and types of drums, however the first compaction is carried out by the paver. A vibrating and heated board (table) levels and pre-compacts the mix and shapes longitudinal and transverse slopes. This is a very important process as it allows to achieve about 90% of the required compaction of the pavement layer. In order to e ffectively develop the road surface design, the compaction of the mixture in the laboratory must properly simulate the compaction in real conditions [25].

According to the European standard, the required level of density expressed as percentage refers to the volumetric density obtained for samples made in the Marshall compactor compared to the results obtained for samples cored from the pavement. The degree of compaction of the layer on site, in addition to the testing of the cores, can also be determined by nuclear and non-nuclear density gauges [26]. In relation to these types of density tests, Micaelo et al. noted that the nuclear method has greater fluctuations in results [27]. Another method may be the density measurement based on an analyzer using neural networks which has the ability to predict the density in real time during the compaction process [28,29].

During road construction (also in laboratory) there is a risk of improper layer compaction, which results from several factors: inappropriate choice of rollers, weather conditions, properties, and temperature of the asphalt mix. Praticò et al. also stated that volumetric properties are also influenced by the test method and even the diameter of the samples [30].

As shown in the presented literature analysis, relevant compaction significantly affects the quality of the asphalt road surface. Moreover, literature data clearly indicate that there are strong correlations between the compaction method and the mechanical properties of asphalt mix. The main aim of this article is to assess the influence of improper compaction (including under or over-compaction) on the quality and durability parameters of the road surface such as density, air void content, stiffness modulus, indirect tensile strength, and resistance to water and frost.
