Product Design Analysis with Regard to Recycling and Selected Mechanical Properties

Abstract

The design process is a complex task in which different goals and properties have to be achieved. Nowadays, end-of-life issues are increasingly being considered in addition to typical design properties such as durability, appearance or quality. This article presents the product design process in relation to its recycling and mechanical properties. A plate connection in two design versions was chosen as the product: in the form of a multi-bolted connection and a multi-riveted connection. An analysis was conducted for several variants of these connections. Recycling properties were considered using various measures calculated from the Recycling Product Model, a type of product model that includes its recycling properties. Selected mechanical properties were determined using the Finite Element Method. Removing one bolt from the connection resulted in a stiffness reduction of almost 11%, while removing two bolts from the connection resulted in a stiffness reduction of almost 26%. In contrast, the removal of one rivet from the connection led to a stiffness reduction of about 3%, while the removal of two rivets from the connection led to a stiffness reduction of less than 5%.

1. Introduction

Product design is a multi-faceted process [1], which is particularly relevant for complex products used in mechanical and equipment engineering [2,3]. Among the typical design, manufacturing or operational aspects [4,5,6,7,8], one can also point to the recycling of the parts that constitute a product at the end of its life [9,10,11].

Growing environmental requirements are playing an increasingly important role in product design, which is represented by the need to consider environmental factors during design [12]. The aim is to create a circular economy in which waste is minimised and resource use is maximised [13]. A crucial part of the circular economy is the product design stage [14,15], where decisions are made to determine a product’s sustainability potential, recyclability and overall environmental footprint. The most important elements associated with the circular economy include design for sustainability, material selection, design for dismantling and reuse and remanufacturing [16,17].

Fundamental to the concept of reuse is the reintroduction of products into the market for their original purpose or through reuse [18]. This approach is consistent with a circular economy, focusing on preserving the value and usefulness of a product after its initial use [15,19]. Significant aspects of reuse are reducing premature disposal, contributing to waste reduction and promoting responsible consumption [20]. The reuse process includes the systematic refurbishment and restoration of used products to their original state or even beyond their original properties [21,22]. It typically involves dismantling, replacement of used parts and rigorous quality control measures. Remanufacturing not only extends the life-cycle of the product but also reduces the need for new materials, minimising the environmental impact related to manufacturing processes [23]. When designing for dismantling, fasteners are one of the points that have the highest impact on product dismantling [24].

One of the primary means of combining parts in products is through multi-bolted connections. Depending on how the assembly is realised, the strength of such connections can vary significantly [25,26,27]. Therefore, the variation of such connections should be an important stage in their design. The second group of connections used in product construction are riveted connections. The choice of assembly techniques at the design stage is becoming increasingly important for the sustainability of high-purity metal recycling due to current recycling practices [28]. Studies on the limitations of aluminium recycling mainly focus on the performance of current sorting and separation processes [29] and the challenges at the metallurgical recycling stage [30]. There is a lack of understanding of the impact of assembly selection during initial product design on the quality of the processed aluminium alloy that is recovered from highly complex products [31].

With regard to the recycling of multi-bolted and multi-riveted connections and products with such connections, several articles have recently been published. Kreilis and Zeltins [32] described the behaviour of reused structural elements with renewed bolted connections under cyclic and ultimate load. The use of bolted connections for the assembly, dismantling and reuse of lightweight external walls was described by Kitayama and Iuorio [33]. Liu et al. [34] conducted three tests on the dismantling of a full-size space frame structure with bolt–ball connections. They found that the recycling rates of structural parts cycles were 100%, 98.62% and 92.9%, respectively. After three cycles, the failure of high-strength bolts used accounted for 16.56%. Dai et al. [35] proposed a system of demountable beam-to-beam connection to enable the reuse of its parts. Soo et al. [36] analysed the impact of connection technologies on the recyclability of end-of-life vehicles, taking into account the screwed and bolted connections found in them.

The topic of recycling aluminium alloy automotive sheets connected by self-piercing rivets has been described by Abe et al. [37], Hoang et al. [38] and in a review by Mori and Abe [39]. Although the steel rivets used in conventional self-piercing riveting are removed from aluminium alloy sheets during recycling, removal is not required for aluminium alloy rivets. The connected sheets with rivets are directly melted due to the same material—aluminium. A life-cycle assessment model and a life-cycle cost assessment model for car parts connected by techniques such as welding, bonding and riveting are presented in [40]. Here, the environmental and cost impacts of the chosen technique were investigated based on energy consumption and material inflows and outflows. A multi-objective decision-making algorithm was also proposed to identify the most suitable techniques for automotive applications. Optimisation of the roof edge profile recycling process using shape memory alloy connecting elements is described in [41].

The issue of analysing product design from a recycling point of view is reflected in the definition of measures proposed in numerous articles. Sodhi et al. [42] suggested the Unfastening Effort Index (UFI), which expresses the effort required to disconnect a particular type of fastener. The disconnection time of a fastener can also be estimated from the UFI value. Another formula for estimating the disconnection time was defined by Vanegas et al. [43]. They distinguish between six categories of dismantling task and sum the dismantling time for each category for a final estimate of the dismantling time. De Aguiar et al. [44] proposed a set of measures to describe product design from dismantling and material properties. The product dismantling properties are assessed in this approach using the following measures: Quantity of Fasteners Index (QFI), Percentage of Fasteners Index (%FI), Type of Fastener Index (TFI), Quantity of Types of Fasteners Index (QFTI) and Accessibility Index (AI). In turn, the product material properties are described by measures, such as Infrastructure Index (II), Material Compatibility Index (MCI), Material Group Index (MGI) and End-of-Life Contamination Index (EoLCI).

In contrast to singular assessment measures, aggregate assessment measures can also be found in the literature. Dostatni [45] and Dostatni et al. [46] introduced the Total Recycling Rate (TRR), which is the sum of three quantities: Material Diversity (MD), Connection Diversity (CD) and Recycling Target (RT). Leal et al. [47] defined an index of Product Recyclability, which concentrates on recycling properties of the materials used in the product design. This aggregated measure includes compatibility of materials, diversity of materials, recyclability of materials and number of aggregated action levers.

This article addresses the issue of product variants [17] and its effects on the recycling process side of the product and on the strength side of the product. The combination of these two aspects has not yet been used in the analysis of connections with fasteners. As an objective, the article set out to investigate how changing the number of fasteners in a connection affects the recycling and mechanical properties of the connection and whether these properties change in parallel. An example of a connection made once as a multi-bolted connection and once as a multi-riveted connection was chosen as the product to be considered. Spatial modelling software (Computer-Aided Design (CAD) type) [48,49] and Finite Element Method (FEM) calculations [50,51] were used as engineering tools in the preparation of the article.

2. Materials and Methods

2.1. Characteristics of the Tested Connections

The assemblies tested consisted of a pair of plates connected by M10 bolts (ISO 4014 M10 × 80 [52]) fastened with clearance and corresponding nuts (ISO 4032 M10 [53]) or 10 mm-diameter round head rivets (type 10 × 65 × 56 [54]) fastened without clearance [55,56]. The connected plates were welded to the top and bottom bases along the two edges of their contact. The thickness of the connected plates and bases was 28 mm. The connections were inclined to the horizontal plane at an angle of 60 degrees [57]. The overall height of each assembly was approximately 266 mm. The connection cases analysed, which differed in the number of fasteners, are shown in Figure 1 and Figure 2. Solid models of the fasteners are illustrated in Figure 3. The basic dimensions of the connection are shown in Figure 4.

The top plate (referred to as plate_vertical_01 in the recycling model) and top base (referred to as plate_up in the recycling model) were made of A356.0-T6 aluminium alloy [58], while the bottom plate (referred to as plate_vertical_02 in the recycling model) and bottom base (referred to as plate_down in the recycling model) were made of A992 steel [59]. Bolts, nuts and rivets were made from 440C stainless steel [60].

2.2. Recycling Product Model

Considering recycling issues at the product design stage requires dealing with data that are not included in a typical 3D CAD model. The data encompass the definition of the connections used in the product and additional material attributes that describe their recycling properties. In order to overcome the limitations of typical 3D CAD systems in terms of product modelling with regard to recycling analysis, this study uses the Recycling Product Model (RPM) [42]. The RPM is an extension of the 3D CAD assembly model and consists, in addition to the 3D CAD assembly model itself, of the following elements:

• Enhanced connection attributes with dismantling attributes (connection constraints);
• Material recycling attributes;
• Product categorisation.

Connection constraints define the connections in the product model from a functional, rather than geometric, point of view, as in a typical 3D CAD system. Material recycling attributes extend the 3D CAD system’s material library and define the attributes that are important from a recycling perspective. Product categorisation is necessary for the calculation of some recycling assessment measures. These elements enable the automatic calculation of recycling assessment measures during the design process. Practical usage of the RPM is possible using the add-in running in the Autodesk Inventor 3D CAD system (Autodesk, San Francisco, CA, USA). Details of RPM are presented in [17].

For the purpose of this study, the RPM was defined for all variants shown in Figure 1 and Figure 2. It involved, based on 3D CAD assembly models, the definition of connection constraints and the addition of material recycling attributes. It was assumed that all multi-bolted and multi-riveted connections could be fully dismantled. Dismantling times were estimated based on a set of measurements performed on original or highly-similar-to-the-original connection types. The connection constraint definitions for Variant 1a and Variant 2a are presented in

Table 1. Connection constraints and dismantling costs of the model in Variant 1a.

ID	Connected Parts	Connecting Parts	Connection Nature	Connection Type	Dismantled	Dismantling Time [s]
cc01	plate_vertical_01/	ISO 4014 M10 × 80:1	Temporary	Threaded	Yes	5
	plate_vertical_02	ISO 4032 M10:1
cc02	plate_vertical_01/	ISO 4014 M10 × 80:2	Temporary	Threaded	Yes	5
	plate_vertical_02	ISO 4032 M10:2
cc03	plate_vertical_01/	ISO 4014 M10 × 80:3	Temporary	Threaded	Yes	5
	plate_vertical_02	ISO 4032 M10:3
cc04	plate_vertical_01/	ISO 4014 M10 × 80:4	Temporary	Threaded	Yes	5
	plate_vertical_02	ISO 4032 M10:4
cc05	plate_vertical_01/	ISO 4014 M10 × 80:5	Temporary	Threaded	Yes	5
	plate_vertical_02	ISO 4032 M10:5
cc06	plate_vertical_01/	ISO 4014 M10 × 80:6	Temporary	Threaded	Yes	5
	plate_vertical_02	ISO 4032 M10:6
cc07	plate_vertical_01/	ISO 4014 M10 × 80:7	Temporary	Threaded	Yes	5
	plate_vertical_02	ISO 4032 M10:7
cc08 1	plate_vertical_01/plate_up	–	Permanent	Welded	No	–
cc09 1	plate_vertical_01/plate_up	–	Permanent	Welded	No	–
cc10 2	plate_vertical_02/plate_down	–	Permanent	Welded	No	–
cc11 2	plate_vertical_02/plate_down	–	Permanent	Welded	No	–
					Total:	35

¹—Connection constraints at the welds on the two edges between the aluminium plates; ²—connection constraints at the welds on the two edges between the steel plates.

and

Table 2. Connection constraints and dismantling costs of the model in Variant 2a.

ID	Connected Parts	Connecting Part	Connection Nature	Connection Type	Dismantled	Dismantling Time [s]
cc01	plate_vertical_01/ plate_vertical_02	PN-M-82952-10 × 65 × 56:1	Permanent	Riveted	Yes	11
cc02	plate_vertical_01/ plate_vertical_02	PN-M-82952-10 × 65 × 56:2	Permanent	Riveted	Yes	11
cc03	plate_vertical_01/ plate_vertical_02	PN-M-82952-10 × 65 × 56:3	Permanent	Riveted	Yes	11
cc04	plate_vertical_01/ plate_vertical_02	PN-M-82952-10 × 65 × 56:4	Permanent	Riveted	Yes	11
cc05	plate_vertical_01/ plate_vertical_02	PN-M-82952-10 × 65 × 56:5	Permanent	Riveted	Yes	11
cc06	plate_vertical_01/ plate_vertical_02	PN-M-82952-10 × 65 × 56:6	Permanent	Riveted	Yes	11
cc07	plate_vertical_01/ plate_vertical_02	PN-M-82952-10 × 65 × 56:7	Permanent	Riveted	Yes	11
cc08 1	plate_vertical_01/plate_up	–	Permanent	Welded	No	–
cc09 1	plate_vertical_01/plate_up	–	Permanent	Welded	No	–
cc10 2	plate_vertical_02/plate_down	–	Permanent	Welded	No	–
cc11 2	plate_vertical_02/plate_down	–	Permanent	Welded	No	–
					Total:	77

¹—Connection constraints at the welds on the two edges between the aluminium plates; ²—connection constraints at the welds on the two edges between the steel plates.

and Figure 5.

TFI, the sum of individual TFIs (understood as an index Total TFI) and QFTI were used as measures to assess the design variant of the products.

TFI represents the difficulty of dismantling of a fastener assessed by a number between 1 and 4. The TFI value is derived from studies on the dismantling effort of particular fastener types [44]. Thus, the sum of the TFI values (i.e., Total TFI) represents the overall difficulty of dismantling the product. The Total TFI value can be calculated using the formula:

$$\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{T}\mathrm{F}\mathrm{I}=\sum _{i=1}^n{\mathrm{T}\mathrm{F}\mathrm{I}}_i$$

(1)

where i is the number of connections used in the product, i = 1, 2, 3, …, n; TFI_i is the TFI value for connection i.

QFTI is specified as the number of different fastener types, limited to 4 [44]. The QTFI identifies the most problematic parts from dismantling in terms of recycling. The choice of the measures presented is based on previous research as the most useful for assessing the recycling properties of products [17].

The calculation of the total recycling profit can be carried out according to the relationship:

$$\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{y}\mathrm{c}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{f}\mathrm{i}\mathrm{t}=\sum \mathrm{r}\mathrm{e}\mathrm{c}\mathrm{y}\mathrm{c}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{f}\mathrm{i}\mathrm{t}-\sum \mathrm{r}\mathrm{e}\mathrm{c}\mathrm{y}\mathrm{c}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{c}\mathrm{o}\mathrm{s}\mathrm{t}$$

(2)

whereby recycling profit is the profit from the sale of product parts; recycling cost is the cost of dismantling product parts. Total recycling profit is a modified form of the formula previously proposed by Karwasz [61].

The recycling profit of the part can be calculated as:

$$\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{y}\mathrm{c}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{f}\mathrm{i}\mathrm{t}=m_p·c_r$$

(3)

where m_p is the mass of the part [kg]; c_r is the unit material recycling profit [currency unit (c.u.)/kg].

The recycling cost can be computed as:

$$\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{y}\mathrm{c}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{c}\mathrm{o}\mathrm{s}\mathrm{t}=t·l$$

(4)

where t is the dismantling time [s]; l is the unit labour cost [c.u./h].

The unit profit from recycling the materials of which the product under assessment is composed was taken from the recycling company [62]. The unit labour cost was determined according to the minimum remuneration per hour in Poland [63]. All monetary amounts were expressed in Polish currency unit—the Polish zloty (PLN). Recycling profit and recycling cost were calculated in accordance with the unit material recycling profit (as part of the material recycling attributes) and the unit labour cost entered into the RPM add-in (Figure 6).

2.3. Finite Element Models

FEM calculations were carried out using Midas NFX 2023 R1 software (MIDASoft, Inc., New York, NY, USA). The parts of the models were assigned the materials indicated in Section 2.1, the properties of which are summarised in

Table 3. Material properties of the connection models.

Part Name	Material	Elastic Modulus [GPa]	Poisson’s Ratio [–]	Tensile Strength [MPa]
Top plate	A356.0-T6	72	0.33	200
Bottom plate	A992	200	0.3	450
Top base	A356.0-T6	72	0.33	200
Bottom base	A992	200	0.3	450
Bolt	440C	200	0.28	758
Nut	440C	200	0.28	758
Rivet	440C	200	0.28	758

.

Hybrid Mesher was used to build the meshes of the individual models. To improve the quality of the contact connections between the parts to be connected, a mesh of different densities was used for the individual parts. However, in no case was the side length of the finite element greater than 7 mm. Mesh parameters for the adopted connection variants are shown in

Table 4. Mesh parameters for the adopted connection variants.

Variant	Number of Nodes	Number of Finite Elements
1a	32,837	47,225
1b	29,762	42,201
1c	26,312	36,494
2a	23,504	30,780
2b	21,590	27,776
2c	19,762	25,150

.

As the subject of the research in this article was not a detailed analysis of the changes in bolt and rivet forces during connection preloading and in the operating condition, no fastener preload was introduced in the models. In addition, the exact value of the clamping stresses in riveted connections is usually unknown [64], and experimental tests would have to be performed to find it out [65]. However, it should be emphasised that the clamping stresses in rivets are not as significant as in the case of preloaded bolts. The clamping of the bolts was modelled by adding contact connections between the bolt heads and the top plate and between the nuts and the bottom plate to the models. Similarly, rivet clamping was modelled by adding contact connections between the respective rivet heads and plates (top and bottom) to the models.

‘General’ contact elements were used between the top plate and the bottom plate, while ‘Welded’ contact elements were used between the other pairs of parts to be connected [66]. These other pairs of parts include the following ones: top plate—top base; bottom plate—bottom base; bolt head—top plate; nut—bottom plate; bolt—nut; rivet head—top plate; rivet head—bottom plate; and rivet shanks—connected plates.

The following values of the ‘General’ contact layer parameters were adopted [67,68]:

• Normal stiffness factor equal to 10;
• Tangential stiffness factor equal to 1;
• Coefficient of static friction equal to 0.14.

Due to the use of ‘Welded’ contact elements, it was possible to prevent the parts from moving relative to each other in any direction [68].

The selected discrete models of the connection according to Variant 1a and Variant 1b are shown in Figure 7a and Figure 7b, respectively. The figures also show how the models were restrained and loaded. In all cases, the models were restrained by taking all degrees of freedom from the underside of the bottom plate and loaded from the top with a uniform pressure p of 0.2 MPa applied in 2 steps, which for a pressure area of 156 cm² corresponds to a working load of 3.12 kN. The Nonlinear Static Analysis module with the geometric non-linearity option [66] was used to perform the calculations.

3. Results

3.1. Product Recycling Properties

An analysis of the recycling properties of the product variants was conducted based on Recycling Product Model using the RPM add-in.

Two measures were used from a product design perspective: QTFI and Total TFI. The QTFI measure indicates two parts as the most problematic from the dismantling aspect of the recycling: plate_vertical_01 and plate_vertical_02. The reason for this is the numerous connections of these parts to other parts of the product. The QTFI values for Variants 1a to 2c are presented in

Table 5. QTFI values for Variants 1a to 2c.

Part Name	Variant 1a	Variant 1b	Variant 1c	Variant 2a	Variant 2b	Variant 2c
plate_vertical_01	2	2	2	2	2	2
plate_vertical_02	2	2	2	2	2	2
plate_up	1	1	1	1	1	1
plate_down	1	1	1	1	1	1

and Figure 8.

Total TFI shows differences in the overall difficulty of dismantling the product. Variant 2a is the most problematic, and this measure clearly indicates this. On the other hand, Variant 1c is the least problematic, and this is also reflected in the Total TFI value. The Total TFI values for Variants 1a to 2c are presented in

Table 6. Total TFI values for Variants 1a to 2c.

	Variant 1a	Variant 1b	Variant 1c	Variant 2a	Variant 2b	Variant 2c
Total TFI	27.2	25.6	24.0	44.0	40.0	36.0

and Figure 9.

The difficulty in dismantling the product variants is also visible in the total dismantling time. The total dismantling time for the worst variant (Variant 2a) is three times longer than for the best variant (

Table 7. Parameters of the dismantling process.

Variant	Total Disassembly Time [s]	Total Disassembly Cost [c.u.]	Material Recycling Profit [c.u.]	Total Recycling Profit [c.u.]
1a	35	0.27	21.60	21.33
1b	30	0.23	21.60	21.36
1c	25	0.20	21.59	21.40
2a	77	0.60	21.53	20.93
2b	66	0.52	21.54	21.02
2c	55	0.43	21.55	21.12

; Figure 10), which directly affects the dismantling costs (Table 7; Figure 11).

The material recycling profits are similar for all variants (Table 7; Figure 12), which is also caused by the differences in mass between the variants. The lack of holes increases the mass of Variants 1c and 2c, which is reflected in an increase in part material revenue.

The total recycling profits are also similar for all variants (Table 7; Figure 13), which is due to the dominant effect of part mass on total recycling profits.

3.2. FEM Calculation Results

The following parameters were used as criteria for comparing the calculation results of the various connection variants:

• Average deflection in the z-axis of the top plate under pressure p—d_tp;
• Stiffness of the connection—k;
• Maximum reduced stresses according to the von Mises formula—s_r.

The average deflection in the z-axis of the top plate under pressure was calculated as the arithmetic mean of the deflections at all nodes lying in the upper part of the top plate.

The stiffness of the connection was determined from the relationship:

$$k={\displaystyle \frac{p·A}{d_{tp}}}$$

(5)

where A is the area of the top plate to which the pressure p was applied (A was 156 cm²).

Parameter values determined for the various connection variants are shown in

Table 8. Selected mechanical properties of the connections.

Variant	dtp [mm]	k [kN/mm]	sr [MPa]
1a	0.1429	21.83	404.6
1b	0.1596	19.55	476.4
1c	0.1922	16.23	621.4
2a	0.0226	137.9	23.45
2b	0.0233	133.7	30.93
2c	0.0236	132.0	36.87

. On the basis of these, the generally more favourable mechanical properties of multi-riveted connections compared to multi-bolted connections can be confirmed, as also demonstrated, for example, in [69]. As can be seen from Table 8 and Figure 14, Figure 15 and Figure 16, multi-riveted connections could be loaded under a much higher working load without failure. However, in order to be able to compare the connection variants tested, the load was limited to one that does not yet induce a yield stress for the bolt and nut material.

The removal of further fasteners (both bolts and rivets) from the connection has the effect of reducing the stiffness of the connection. A greater decrease in this stiffness was observed for multi-bolted connections.

The highest values of reduced stresses occurred in the fasteners, but in neither case did they exceed the yield strength of the material from which they were made. Example reduced stress maps for the fasteners in the case of connection according to Variant 1a and 2a are shown in Figure 17a and Figure 17b, respectively.

4. Discussion

As the number of fasteners of a given type decreased, the difficulty of dismantling the product design increased, which is reflected in the values of the Total TFI evaluation measure. However, the difficulty of disassembly should be evaluated in conjunction with other recycling assessment measures such as total recycling profit. In the product design variants analysed, the worst variant (Variant 2a) was 2% less profitable than the best variant (Variant 1a), as a result of the relationship between material recycling profits and product dismantling costs.

On the other hand, the opposite trend can be observed with regard to product performance. Due to the clearances occurring between the bolts and the holes in the plates to be connected, the stiffness of multi-bolted connections was significantly lower than that of corresponding multi-riveted connections. In the case of connections with seven fasteners, the relative difference in the stiffness of the connections was almost 84% and increased even slightly as the number of fasteners was reduced. Similarly, the reduced stresses occurring in multi-bolted connections reached higher values than those observed in multi-riveted connections.

It should also be noted that, as the number of fasteners, both bolts and rivets, decreased, the mechanical properties of the connection deteriorated. Removing one bolt from the connection resulted in a stiffness reduction of almost 11%, while removing two bolts from the connection resulted in a stiffness reduction of almost 26%. In contrast, in the case of riveted connections, the removal of one rivet from the connection led to a stiffness reduction of about 3%, while the removal of two rivets from the connection led to a stiffness reduction of less than 5%.

To combine the advantages of multi-bolted and multi-riveted connections, one could consider creating a hybrid connection that uses both types of fasteners (bolts and rivets). An attempt to analyse such connections was made, for example, in [70]. In contrast, when considering the need to melt aluminium alloy and steel during the recycling process together (if the connection could not be dismantled), it would be necessary to investigate how impurities would affect the mechanical properties of the aluminium alloy after recycling [71].

Two limitations of the presented study can be identified. The first limitation concerns the omission of factors such as the effect of fastener preload on connection strength [72,73] or fatigue caused by external loads [74]. The second limitation concerns the disassembly of the rivets. This study assumes that material loss during disassembly of this type of connection is negligible, due to the relationship between the masses of the connected and connecting parts. These limitations will provide directions for future research.

5. Conclusions

It can be concluded from the study that as the number of fasteners, both bolts and rivets, decreases, the overall dismantling difficulty of the product increases. The overall dismantling difficulty should be analysed with regard to recycling profit. The relationship between overall dismantling difficulty and total recycling profit depends on the mass of the product parts.

During finite element testing, it was shown that the stiffness of multi-bolted connections is significantly lower than that of the corresponding multi-riveted connections, due to the clearances that occur between the bolts and the holes in the interconnected plates, while the rivets fit tightly into the plate holes. It was also observed that as the number of fasteners, both bolts and rivets, decreases, the mechanical properties of the connection deteriorate.

As a further development, it is proposed that optimisation of the fastener position is included in the considerations to improve the mechanical properties of the connection.