An analytic vehicle simulation method is used to calculate gasoline ICEV and BEV energy use and mass [5]. In this approach driving cycle dependent coefficients are combined with vehicle resistance characteristics, i.e. the frontal area, aerodynamic drag, tire rolling resistance, and vehicle mass to calculate mechanical energy demand [11]. To convert vehicle mechanical energy demand for a certain driving cycle into vehicle energy use, driving cycle averaged power train efficiencies are calculated using Advisor, a numeric vehicle simulation software [12]. The calculation of vehicle energy demand and average operating point efficiencies is based on the New European Driving Cycle (NEDC), the standard driving cycle for emission certification in Europe. Several studies have shown that the fuel consumption as measured by vehicle manufacturers for the NEDC driving cycle is significantly lower than “real-world” consumption and that this discrepancy increased over time to about 20% in 2012 [13]. Reasons for this include among other things the increasing use of tolerances in the determination of road load, test temperatures, and transmission shifting schedules for type-approval, as well as the increasing market share of vehicles equipped with air conditioning systems [13]. These causes do not apply to the simulation presented in this work since the above factors are assumed to correspond to real driving conditions. If generally a more aggressive drive cycle than the NEDC is deemed appropriate, the results for energy use must be scaled accordingly. This would result in a higher share of fuel or electricity to total vehicle costs and accordingly a higher degree of light weighting being optimal. An analytic expression is used to evaluate vehicle mass and energy use as a function of primary mass reduction, vehicle configuration parameters such as vehicle range and performance, and technical parameters such as battery specific energy [5]. The used method allows keeping vehicle range and performance constant while reducing vehicle weight. The baseline vehicle configuration corresponds to a midsize passenger car (see Table 1).

Vehicle m_veh is calculated as the sum of power train mass m_pt (engine system for the ICEV and electric motor including controller for the BEV), the energy storage mass m_es (fuel tank including fuel for the ICEV and Li-ion battery for the BEV), glider mass m_gl, (the glider represents all remaining vehicle parts, i.e. the body, chassis, interior, etc.) and additional material m_sup necessary for structural support of the powertrain and energy storage beyond the glider baseline (this is particularly relevant for structural support of the battery in a BEV):

The baseline glider is based on an inventory of the 2008 Mercedes Benz A-Class [14], [15]. The glider mass consists to 72% of steel and to 28% of other materials. It is assumed that light weighting is applied to the part consisting of steel (primarily used in the body, doors, and chassis). The part made of materials other than steel (mainly found in the vehicle interior, tires, powertrain fluids, etc.) is kept constant. The base glider mass is slightly higher for the ICEV than for the BEV due to additional mass for the transmission and exhaust system. Powertrain and energy storage mass and cost are calculated as the sum of a fixed amount and a fraction that scales linearly with power and energy storage capacity. Table 2 lists relevant component-specific mass and cost baseline assumptions. Note that these numbers represent the current technology status. Evaluation of light weighting effects for future scenarios is not analysed in this paper, however the sensitivity to important input parameters, in particular battery specific energy and cost, is analysed.

Manufacturing costs are calculated as the sum of powertrain, energy storage and glider cost. Glider cost C_gl is the sum of a fixed cost for the part that is not light weighted (corresponding to m_gl,fix) and a variable cost that is dependent on the amount of weight reduction implemented (corresponding to m_gl,var).

Glider specific fixed cost is SC_gl,fix = 14.3 $/kg and variable baseline (without use of lightweight materials) cost is SC_gl,var,0 = 5.7 $/kg. Both specific cost values are based on the mass and manufacturing cost breakdown for a midsize passenger car given in [16]. Specific cost scaling is based on a literature review of the manufacturing cost increase of lightweight materials relative to steel. In order to analytically minimize vehicle cost a continuous lightweight cost function is used. This concept is rather artificial as in practice discrete material options lead to discrete weight reduction potentials and costs. However for the purpose of this analysis, i.e. the comparison of the cost reduction effects of light weighting for different power train types, it is useful. Figure 1 shows the literature data [21]-[25] together with a fit for a quadratic polynomial with an asymptote at the assumed maximum amount of light weighting Δm_gl,var,max which is assumed to be 80%.

Figure 1.

Light weighting manufacturing cost increase relative to steel. Shown are literature data (circles) and a low, average, and high light weighting cost function

The “average cost” function in Figure 1 corresponds to a least squares fit which yields a = 5.5. It will be referred to as the baseline light weighting cost function in the following analysis. The data and fit shows the general trend that the more mass the lightweight material can substitute relative to steel, the more expensive it is to manufacture (per unit of weight). Reductions of up to 20-30% can be achieved at relatively low cost by substituting steel with high-strength steel. Higher reductions require materials with a higher amount of labour and energy input per kg of material, e.g. carbon fibre. Due to the spread and limited availability of data, two additional cost functions for a lower (a = 2.75) and a higher (a = 11) bound are defined.

To convert manufacturing costs to retail price, a mark up factor of 1.4 is used [16]. Total costs are calculated as the sum of vehicle purchase and lifetime energy costs. Maintenance and repair costs are assumed to be independent of light weighting and are therefore not considered. For simplicity no discounting, i.e. devaluation of future energy costs, is used. In the base case a lifetime of ten years, an annual driving distance of 15,000 km, a gasoline price of 2 $/L and an electricity charging cost of 0.35 $/kWh are assumed. Charging cost is based on an electricity price of 0.2 $/kWh, an energy based tax of 25 $/GJ, and 0.06 $/kWh charging station cost. For the BEV the total driving distance of 150,000 vehicle-kilometre (km_v) is limited by the lifetime of the battery. For the ICEV a second base case with 300,000 km_v is analysed.

The cost minimum is found by setting the partial derivative of manufacturing and total costs to zero and solving for the optimal degree of light weighting as a function of the parameters of interest. The explicit equations for the optimal degree of light weighting are not given as they are generally rather long, instead the results are shown graphically.

Figures 2a and 2b show total ICEV and BEV mass as a function of variable glider mass reduction. Even though the base glider mass is slightly higher for the ICEV than the BEV, total vehicle mass is higher for the BEV due to the additional mass of the battery. As the weight of the glider is reduced, the sizes of the powertrain and energy storage decrease if acceleration performance and range are kept constant. This effect is referred to as secondary weight reduction in the following. Figure 2c shows vehicle energy use for the NEDC as a function of variable glider mass reduction. Due to the lower powertrain efficiency and the lack of regeneration capability, the sensitivity of energy consumption to mass reduction is lower for the BEV than the ICEV. Note also that the sensitivity of energy consumption to mass reduction is generally higher in urban driving conditions, due to a higher share of kinetic energy (for acceleration) to total mechanical energy demand. This makes light weighting particularly useful in urban driving conditions, e.g. for urban delivery vehicles.

Figure 2.

Vehicle mass (a, b) and energy consumption (c) for the ICEV and BEV (200 km range) as a function of variable glider mass reduction

The secondary weight reduction effect is particularly relevant for the BEV due to the high mass and cost of the battery. It is reflected in the total-to-primary weight reduction, i.e. the ratio of total vehicle mass reduction (including secondary effects) to primary glider mass reduction, which equates in the baseline case to 126% for the BEV (at 200 km range) and 111% for the ICEV. Secondary weight reduction effects in a BEV are particularly dependent on the battery size and as such on vehicle range and battery specific energy. Figure 3 shows the total-to-primary vehicle weight reduction as a function of electric range and battery specific energy. It can be seen that secondary weight reduction is significant for current battery technology (ca. 100 Wh of usable energy per kg battery pack) at ranges above ca. 200 km. For future batteries reaching 300 Wh/kg secondary weight effects become significant only above ca. 600 km range.

Figure 3.

Total-to-primary BEV weight reduction as a function of vehicle electric range and battery specific energy

Note that the assessment in this paper does only take into account secondary weight reduction effects within the powertrain, energy storage, and support structure but not within the glider itself, as for example investigated in [26]. In principal the method can be easily adapted to include this effect. If secondary weight reduction within the glider were included, total secondary effects would be higher and the results shown in the following slightly more positive towards the use of light weighting. However, the general conclusions regarding the relative benefits of light weighting in conventional and electric vehicles remain unchanged.

Figure 4a shows ICEV manufacturing cost as a function of variable glider mass reduction for the baseline lightweight cost function. Minimum manufacturing cost is reached at ca. 10% weight reduction. Figure 4b and 4c show the breakdown of ICEV total cost also as a function of variable glider mass reduction for the baseline lightweight cost function. It is apparent that the share of fuel costs to total costs is relatively high and dominates if no lightweight material is used. Due to the high share of fuel to total costs and the high sensitivity of fuel consumption to weight reduction, the difference between the optimal weight reductions minimizing manufacturing versus total costs is relatively large and increases further at higher driving distance. For 150,000 km_v (Figure 4b) minimum total cost is reached at 22% and for 300,000 km_v (Figure 4c) at 31% variable glider mass reduction.

Figure 4.

Effect of light weighting on the manufacturing (a) and total costs 150,000 km_v (b); 300,000 kmv (c) for the ICEV. Black points indicate optimal levels of light weighting minimizing manufacturing and total costs

Figure 5a shows the sensitivity of the optimal weight reduction minimizing ICEV total cost relative to lifetime driving distance. The optimal degree of light weighting is very sensitive to the driving distance and the considered light weighting cost function. Figure 5b shows the corresponding total cost reduction. For the baseline lightweight cost function it is $1, $780 and $4,130 for 150,000 and 300,000 km_v, respectively.

Figure 5.

Optimal weight reduction minimizing ICEV total cost as a function of lifetime driving distance for three lightweight cost functions (a); Corresponding total cost reduction (b)

Figure 6a shows the optimal amount of light weighting minimizing total costs as a function of vehicle driving distance and fuel price. In this case the baseline light weighting cost function is assumed. The optimal use of light weighting is very sensitive to both parameters. Figure 6b shows the corresponding total cost reduction.

Figure 6.

Optimal weight reduction (indicated in % in the legend) to minimize total cost of an ICEV as a function of fuel price and km_v (a); Corresponding total cost reduction (in $) (b)

Figure 7a shows BEV manufacturing cost as a function of variable glider mass reduction. In this case the baseline lightweight cost function and battery costs according to Table 2 are assumed. Minimum manufacturing cost is reached at 24% weight reduction. The higher optimal use of light weighting in the manufacturing phase relative to the ICEV (24% vs. 10%) is due to the high cost of the battery. Reducing vehicle weight it can be downsized while maintaining the same driving range which reduces manufacturing costs and results in the higher optimal degree of light weighting relative to the ICEV. Figure 7b and 7c show total cost as a function of weight reduction for a BEV with a range of 200 km and 400 km, respectively. In this case the baseline charging cost of 0.35 $/kWh and a driving distance of 150,000 km_v are assumed. Minimal total cost is reached at 28% and 39% variable glider mass reduction for a BEV range of 200 km and 400 km, respectively. The small difference between the solutions minimizing manufacturing and total costs is due to the low share of electricity to total costs and the relatively low sensitivity of BEV energy consumption to weight reduction.

Figure 7.

BEV manufacturing (a) and total costs (b, c) as a function of variable glider mass reduction. In b) the driving range is 200 km while in c) it is 400 km. Black points indicate optimal levels of light weighting minimizing manufacturing and total costs

The high sensitivity of the optimal degree of light weighting to BEV range is analyzed in Figure 8. Figure 8a shows the sensitivity of the optimal weight reduction minimizing BEV total costs for the three lightweight cost functions considered as a function of BEV range. Figure 8b shows the corresponding total cost reduction. For the baseline lightweight cost function it is $3,070 and $7,620 for a range of 200 and 400 km, respectively. Compared to the analysis shown in Figure 5 for the ICEV, the relation between total cost reduction and range at a given amount of light weighting is nonlinear for the BEV due to secondary scaling effects that occur for an increase in range as explained earlier.

Figure 8.

Optimal weight reduction minimizing total cost of the BEV as a function of vehicle range for three lightweight cost functions (a); Corresponding total cost reduction (b)

Figure 9a shows the optimal amount of light weighting to minimize total costs as a function of vehicle range and battery cost for the baseline lightweight cost function. Figure 9b shows the corresponding reduction in total cost. It is obvious that as battery specific mass and cost decrease, the incentive for light weighting decreases too due to lower associated battery cost and mass reduction potentials.

Figure 9.

Optimal weight reduction (in %) minimizing BEV total cost as a function of specific battery cost and vehicle range (a); Corresponding total cost reduction (in $) (b)