Journal of Cleaner Production 11 (2003) 61–77 www.cleanerproduction.net
Elucidating complex design and management tradeoffs through life cycle design: air intake manifold demonstration project G.A. Keoleian ∗, K. Kar
1
Center for Sustainable Systems, University of Michigan, Dana Building, University of Michigan, Ann Arbor, MI 48109-1115, USA Received 20 June 2001; accepted 3 January 2002
Abstract The life cycle design (LCD) framework for enhancing design analysis and decision making is demonstrated through a collaborative effort between the University of Michigan, a cross functional team at Ford, and the US Environmental Protection Agency. The LCD framework was used to evaluate three air intake manifold designs: a sand cast aluminum, brazed aluminum tubular, and nylon composite. Life cycle inventory, life cycle cost and product/process performance analyses highlighted significant tradeoffs among alternative manifolds, with respect to system design requirements. The life cycle inventory indicated that the sand-cast aluminum manifold consumed the most life cycle energy (1798 MJ) compared to the tubular brazed aluminum (1131 MJ) and nylon composite (928 MJ) manifolds. The cast aluminum manifold generated the least life cycle solid waste of 218 kg per manifold, whereas the brazed aluminum tubular and nylon composite manifolds generated comparable quantities of 418 kg and 391 kg, respectively. Material production accounted for 70% of the total life cycle solid waste for the brazed tubular manifold, while auto shredder residue was responsible for half the total waste for the nylon composite design. The life cycle cost analysis estimated Ford manufacturing costs, customer gasoline costs, and end-of-life management costs. The nylon composite manifold had the highest estimated manufacturing costs but the least use phase gasoline costs. Significant end-of-life management revenues from aluminum recycling would accrue to Ford under automobile take back legislation. A total of 20 performance requirements were used to evaluate each design alternative. 2002 Elsevier Science Ltd. All rights reserved. Keywords: Life cycle design; Life cycle assessment; Life cycle cost analysis; Design decision making; Design requirements
1. Introduction Life cycle design (LCD) was developed to integrate environmental considerations more effectively into the product design process [1-3]. The design of a product is a master blueprint that is a major determinant of its marketplace success as well as its environmental footprint. This ‘footprint’ can be defined by the sum of the environmental burdens and impacts associated with a product’s total life cycle, which encompasses material production, manufacturing, use, and end-of-life management stages. This life cycle provides a comprehensive system boundary for design analysis. Ultimately LCD seeks to enhance design and management decision making by evaluating the full set of performance, cost, pol∗ Corresponding author. Tel.: +1-734-764-3194; fax: +1-734-6475841. E-mail address:
[email protected] (G.A. Keoleian). 1 Present address: Delphi Automotive Systems, Ann Arbor, USA
icy, regulatory and other environmental requirements that influence the product system. A full understanding of the specific requirements that affect the life cycle system offers the greatest potential for achieving more economically and ecologically sustainable product systems. Integration of environmental considerations into the design process represents a complex challenge to product designers, process engineers, product development managers and environmental professionals. Today, successful environmental integration must be achieved within the context of shortening time to market cycles, more stringent regulations, and global competitiveness. A holistic framework including definitions, objectives, principles and tools is essential to guide the development of more ecologically and economically sustainable product systems. In 1991, the US Environmental Protection Agency collaborated with the University of Michigan to develop the life cycle design framework [1-3]. This framework is documented in two publications: Life
0959-6526/02/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 9 5 9 - 6 5 2 6 ( 0 2 ) 0 0 0 0 4 - 5
62
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
Cycle Design Guidance Manual [1] and the Life Cycle Design Framework and Demonstration Projects [3]. The objective of life cycle design is to enhance environmental performance across the life cycle while also optimizing functional performance, cost, and regulatory/policy requirements that influence the product system. Several other investigators have also recognized the value of a multi-objective analysis approach [4-7]. A series of demonstration projects were conducted by the National Pollution Prevention Center (now the Center for Sustainable Systems) through the Cleaner Products Through Life Cycle Design research grant (US EPA: CR822998-01-0). Demonstration projects with industrial partners have targeted a wide variety of products ranging from automotive components to photovoltaics as indicated in Table 1. Three main elements of the design process include the specification of systems requirements, synthesis of strategies for optimizing the design, and evaluation of design alternatives. Specific tools for aiding in the development of requirements include design checklists, guidelines and matrices. Strategies for achieving more sustainable product systems have been classified as product oriented (e.g., enhance durability, performance, reusability, serviceability, remanufacturability, adaptability), process oriented (process substitution, process material and energy efficiency, process control, improve material and inventory control, optimize facility design and layout) and distribution oriented (optimize transportation, reduce packaging, use alternative packaging materials). After synthesizing strategies into feasible design alternatives, these alternatives are evaluated using design analysis tools such as life cycle assessment (LCA). LCA is an analytical tool for quantifying and characterizing the environmental burdens associated with a product life cycle from material production through retirement [22-24]. Life cycle design, by nature of its scope and scale, encompasses a large set of system variables and parameters. For this reason, modeling and framing life cycle design decisions can be a complex process. The process
is further complicated because multiple decision-makers manage this system concurrently. These decision-makers include multi-stakeholders such as manufacturers, suppliers, distributors, customers, regulators, investors, insurers, service managers, resource recovery and waste managers. While most major corporations have developed environmental policies, implementation of these policies at operational level poses significant challenges in product development. The purpose of this paper is to investigate the use of life cycle design methods for improving design analysis and design decision making. A life cycle design demonstration project between the University of Michigan and Ford Motor Company that examined automotive air intake manifolds serves to illustrate the methods. The key features highlighted here are the application of life cycle assessment and life cycle cost analysis methods, and the compilation and evaluation of system requirements, including performance, cost and environmental design criteria. The life cycle assessment and life cycle cost analysis results are interpreted with respect to system requirements, which places the life cycle design analysis in a business context. The combination of life cycle methods and system requirements also serves to elucidate complex tradeoffs among alternatives that exist between environmental categories, such as energy, greenhouse gas emissions, and solid waste, along with performance and cost issues that traditionally dominate decision making.
2. Methods The design analysis of the three alternative air intake manifolds was conducted following the US EPA Life Cycle Design Framework [1,3]. Design analysis conducted in this project focuses on three main elements that are indicated in Table 2. The performance, cost and environmental analyses address production, use, and end-of-life management stages. This paper focuses on system requirements relevant to designing air intake
Table 1 Life cycle design demonstration projects sponsored by the US EPA between 1994–1998 Product system
Industry partner
Fuel Tank Systems: Steel and multi-layer HDPE Instrument Panel
General Motors [8] [9] US EPA Common Sense Initiative: [10] [11] Automotive Sector Ford [12] [13] Ford [14] 3M [15] Ford [16] Ford [17] Dow [18] [19] United Solar [20] [21]
Air Intake Manifold: Al (cast and extruded) and Nylon Lower Plenum of Air Intake Manifold: Al (cast) and Nylon In-Mold Surfacing Film Transmission Case: Cast Al and Mg Transmission End-Housing: Cast Al and Thixomolded Mg Milk and Juice Packaging: Various single use and reuseable containers Amorphous Silicon Photovoltaics
Reference
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
63
Table 2 System requirements and analytical tools used in the air intake manifold life cycle design demonstration project
System Requirements
Performance Analysis
Cost Analysis
Environmental Analysis
쐌 Manufacturability
쐌 Ford costs (manufacturing and warranty costs) 쐌 Gasoline costs 쐌 End-of-life management costs (processing and salvage value) Life cycle cost analysis
쐌 Internal policies
쐌 Operation
Analytical Tools
Engineering testing and modeling
manifolds and two analytical tools for evaluating design alternatives, life cycle inventory analysis and life cycle cost analysis. Ultimately, design decisions are made by evaluating alternatives with respect to system requirements including performance, cost, environmental, and regulatory objectives. 2.1. Product system Three different intake manifolds are investigated: a 2.74 kg glass reinforced nylon composite, a 6.5 kg sandcast aluminum and a 3.62 multi-tube brazed aluminum. The material composition of each manifold is indicated in Table 3. The composite manifold consists of 33% glass reinforced nylon (PA6.6 GF33), brass (UNS C36000) inserts and stainless steel (304 steel) exhaust gas recirculation (EGR) tube. The nylon manifold is manufactured by the lost core process using a tin-bismuth lost core. A stainless steel external EGR tube is required for this design to transfer exhaust gas back into the intake manifold for NOx control. Brass inserts are used in mounting the manifold to the engine block. UNS C36000 brass, which is more commonly known as 360 brass, consists of 77% copper, 20% zinc and 3% lead. 360 brass has a high scrap content and is usually made at the extruder’s facility. In this analysis, 360 brass is assumed to be composed of 99% scrap [25]. 304 stainless steel is made from 100% scrap [26]. The sand-cast aluminum manifold consists of 100% secondary aluminum. The multi-tube brazed aluminum manifold consists of four bent extruded runners and an
쐌 External policies and regulations
Life cycle assessment
extruded plenum screwed to the engine block through a sand-cast flange. The sand-cast flange section comprises 65% of the manifold weight; the extruded sections account for the remaining 35%. Material for the sandcast flange section consists of 100% secondary aluminum, whereas the extruded sections are assumed to be made of 70% primary and 30% secondary aluminum [27] which is a representative mix of extruded parts. Thus, overall the multi-tube brazed manifold consists of 24.5% primary aluminum and 75.5% secondary aluminum. 2.2. Life cycle inventory analysis Details of the life cycle inventory analysis, including process flow diagrams for each manifold, modeling methods and assumptions and data sources are provided in a full project report [13]. The life cycle energy, solid waste, air emissions, and water effluents are presented in this paper. 2.2.1. Material production Inventory data for the material production of each manifold were collected from secondary sources. Table 4 indicates the total environmental burdens for the production of nylon, glass fiber, brass and stainless steel materials used to manufacture the composite manifold. The sand cast manifold and the multi-tube brazed manifold are manufactured using aluminum. The sand cast manifold is produced from secondary aluminum exclusively, while the multi-tube brazed manifold uses both secondary and primary sources. The material pro-
Table 3 Product composition of intake manifolds (mass in kg) Materials
Nylon Composite Manifold
PA6.6 GF33 Stainless steel 304 Brass 360 Secondary aluminum Primary aluminum Total mass
2.07 0.41 0.26
2.74
Sand Cast Manifold
Brazed Al Tubular Manifold
6.5
2.73 0.89 3.62
6.5
64
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
Table 4 Environmental data for materials production of the composite manifold Primary Energy (MJ/IMa) Waste (g/M) Air emissions Carbon dioxide Particulates Nitrogen oxide Sulfur dioxide Carbon monoxide Hydrocarbon Methane Fluorine Hydrochloric acid Heavy metals Halogenated hydrocarbon Solid waste Water effluents Dissolved solids BOD COD Suspended solids Acids Heavy metals Oils Nitrates Chlorides Water (l) Halogenated hydrocarbon a
297
8530.0 16.1 36.0 62.0 23.0 6.0 82.0 1.0 0.5 4×10⫺4 3.1×10⫺3 956.0 701.0 3.0 25.0 116.0 4.0 0.6 1.5 1.6×10⫺2 51.0 20200.0 6.8×10⫺2
IM=intake manifold. Source: [28,29]
duction data for secondary and primary aluminum production are indicated in Table 5. 2.2.2. Manufacturing phase Manufacturing unit processes for the three manifold systems are shown in Table 6. 2.2.3. Use Use phase energy and wastes were calculated for an assumed manifold life of 150,000 miles (241,350 km) in a 1995 Contour. The contribution of the manifold to vehicle fuel consumption, F(l), was obtained using the following correlation: F(l) ⫽ MIM × L ×
冋 册
⌬f FE(l) × Mv ⌬M
where, F(l)=fuel (liters) used over the life of intake manifold (L); MIM=mass of the intake manifold; Mv=test ⌬f =fuel consumpweight (mass) of vehicle=1471 kg, ⌬M tion correlation with mass. For a 1995 Contour the correlation was obtained from the Ford core team as: 10% weight reduction is equivalent to 4% fuel consumption ⌬f =0.4; FE(l)=fuel consumption reduction. Therefore, ⌬M
in liters/km. Fuel economy for a 1995 Contour is 7.46 l/100 km. Therefore, FE=0.0746; L=life of intake manifold=241,350 km. The lifetime fuel consumption and energy for the three manifolds are indicated in Table 7. 2.2.4. End-of-life management The end-of-life management stage of the manifold is defined by processes for the total vehicle which include dismantling, shredding, recycling and waste disposal. Few manifolds are dismantled for reuse, given that they typically do not fail. The inventory model for this stage of the manifold accounted for shredding energy, separation energy and transportation energy. 2.3. Life cycle cost analysis The life cycle cost analysis traces the conventional costs accrued to manufacturers, customers, and end-oflife vehicle managers associated with the air intake manifold. This study did not attempt to measure hidden, contingency (with the exception of warranty), and less tangible costs (e.g., potential increased productivity and revenues associated with environmentally preferable products) as defined by US EPA [35]. For example, special permitting, reporting, tracking and other hidden environmental costs that may be associated with the use of hazardous materials in the manufacturing phase were not analyzed. While a more detailed accounting of these costs would provide more accurate data for decision making, such a total cost assessment was outside of the scope of this life cycle design project. This cost assessment relied on market values from industry or public sources, which often do not reflect the full environmental costs. Since the Contour was marketed in both US and Europe, the cost analysis included a European (German) scenario as well as a US scenario. The objective of this scenario analysis was to explore differences in market conditions that affect the use phase and end-of-life stages of the air intake manifold. The German scenario accounts only for differences in gasoline and landfill disposal costs; no attempt was made to estimate the differences in material costs and manufacturing costs in Germany. Manufacturing costs have two main components: fixed costs (production cost, prototype tooling cost and development cost) and variable costs (wages, salaries, raw materials). Because manufacturing costs were proprietary, indirect cost estimates were used. The variable manufacturing cost of the manifold is estimated as one sixth of the dealer part cost. Thus, Cvar. mfg ⫽
Cdealer 6
The project team also provided the differential manufac-
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
65
Table 5 Environmental data for primary and secondary aluminum production Metrics
Primary Al
Secondary Al
Data Source
163.73 188.40 171.20 170.00 196.30
16.76 13.25 15.60 18.00 26.00
[30]; [31]; [31]; [32]; [33];
Energy (MJ/kg)
Solid waste (kg/kg) alumina production 1 alumina production 2 alumina production 3 average alumina production (red mud) electrolysis cleaning/casting energy smelting energy supply Total Air emissions (kg/kg) CO2 CO SO2 NOx Particulates HC FC HCl H2 Others Water use (m3/kg) Water effluents (kg/kg) Dissolved solids Suspended solids BOD COD Acids Metal ions Lead Tar Fluorides Others
2.0 3.0 2.9 2.63 3.57×10⫺2 2.0×10⫺2 0.27
2.96 13 1.65×10⫺2 9.19×10⫺2∗ 2.85×10⫺2 ∗ 1.96×10⫺2 ∗ 3.77×10⫺3 5.25×10⫺4 1.0×10⫺3 11.44∗
2.55 0.013 1.27 31.47 3.60 0.97 0.003 0.002 0.001 5.77
turing cost between the composite and sand cast manifolds. |Cvar. mfg|comp ⫺ |Cvar. mfg|sc ⫽ $11.50 In addition, warranty costs are direct costs borne by Ford due to product defects or improper assembly of the manifold on the engine. The life cycle costs incurred outside of Ford’s domain during the use and retirement phases represent costs to the customers and vehicle recyclers. In the use phase gasoline costs to the users were evaluated based on results from the life cycle inventory analysis for gasoline consumption. It was assumed that the manifolds perform without maintenance costs to the
4.3×10⫺2 1.87×10⫺2 0.062 0.86 2.21×10⫺4 1.33×10⫺3∗∗ 3.58×10⫺3 ∗∗ 3.57×10⫺4 ∗∗ 2.61×10⫺3 1.3×10⫺3 7.5×10⫺4 5.0×10⫺5 1.6∗∗
German condition Alcoa Worldwide operations Swiss study European study US condition
[27]; estimate Europe [27]; estimate Western Australia [30]; German condition average of alumina 1, 2, 3 production [30]; German condition [30]; German condition [30]; German condition [30]; German condition [30]; German condition Reasonable average condition [31]; Alcoa worldwide operations [30]; Europe condition ∗ [27], ∗∗[30] Europe condition ∗ [27], ∗∗bib:[39] Europe condition ∗ [27], ∗∗[30] Europe condition [30]; German condition [34]; estimated global average [30]; German condition [30]; German condition [30]; German condition ∗[27]; estimate Western Australia∗∗[30]; German condition [30]; German condition
1.1×10⫺4 3.0×10⫺5
1.0×10⫺4
owner over 150,000 miles. The US average cost (Cf) for gasoline was estimated as US $1.24/gallon [36]. The German average cost (Cf) for gasoline was estimated as US $3.34/gallon [36]. The retirement costs were estimated from the retirement spreadsheet model of American Plastics Council [37] and data obtained from several other sources [38,39] [25,30]. This analysis accounted for transportation, processing, salvage values and landfill disposition. In the retirement phase, environmental data evaluated were shredding energy, nonferrous separation energy and transportation energy [37,40]. Emissions and wastes for different life cycle stages were obtained as the sum of process and fuel-related wastes.
66
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
Table 6 Manufacturing unit processes for the three manifold systems Manifold
Component
Manufacturing Unit Process
Composite
Nylon manifold
Brass fittings Stainless steel EGR tube
Sand cast
Multi-tube brazed
Aluminum manifold
Sand-cast aluminum flange
Extruded tubes and air collection chamber
casting tin-bismuth melt cores injection molding —mold and core insertion —overmolding inductive melting tin-bismuth core washing post manifold assembly extrusion machining stamping extrusion brazing green sand preparation mold and core insertion gating and riser preparation melting and pouring post casting machining green sand preparation mold and core insertion gating and riser preparation melting and pouring post casting machining extrusion bending of tubes arrangement brazing
Table 7 Fuel consumption and use phase energy (total fuel cycle which includes upstream activities (e.g., extraction, fuel processing) +combustion) contribution of intake manifolds
Manifold Type
Weight (kg)
Fuel Consumption F(gal), (gallons) F(l), (liters)
Composite manifold Sand-cast manifold Multi-tube brazed manifold
2.74 6.50 3.62
13.4 31.8 17.7
2.4. System requirements Specification of requirements is one of the most critical design functions. Requirements guide designers in translating product functional objectives and other objectives into successful designs. Environmental requirements should focus on minimizing natural resource consumption, energy consumption, waste generation, and human health risks, as well as promoting the sustainability of ecosystems. A primary tool of life cycle design is the multicriteria matrices for specifying requirements. Each matrix is organized by life cycle stages and product system components: product, process and distribution. Elements can then be described and tracked in as much detail as necessary. Requirements can include
3.54 8.40 4.86
Total Fuel Cycle Energy (MJ) 564 1337 745
qualitative criteria as well as quantitative metrics. A simplified version of this matrix is shown in Fig. 1. The matrix allowed product development team members to study the interactions and tradeoffs between environmental, cost, and performance requirements. The project team compiled environmental requirements including government regulations and corporate guidelines and policies. Life cycle costs were grouped according to those direct costs incurred by Ford, including warranty costs and those costs occurring outside Ford’s cost domain. Ford manufacturing costs are a primary criterion for making a business decision. Team members recognized that several requirements were more important in Europe compared to the US, including gasoline costs and vehicle retirement costs. Since Ford is a global company the global perspective was considered.
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
Fig. 1.
Simplified system requirements matrix.
Each manifold alternative must meet basic performance criteria to become a viable candidate for a particular design application. These criteria were identified by the Ford manifold design engineer.
3. Results Life cycle inventory and cost results for the three air intake manifolds are presented here, based on a 150,000 mile service life for a Ford Contour passenger car. The life cycle inventory and cost analyses data are then evaluated and discussed in the context of system requirements. 3.1. Life cycle inventory analysis The life cycle inventory analysis shows some significant environmental tradeoffs among alternative manifold designs. These tradeoffs are apparent when comparing energy, waste, and material profiles. 3.1.1. Life cycle energy The life cycle energy requirements for the three alternative manifolds are shown in Fig. 2. The cast aluminum
Fig. 2.
67
manifold consumes the greatest total life cycle energy (1800 MJ) followed by the brazed aluminum tubular manifold (1130 MJ) and then the nylon composite manifold (930 MJ). The use phase accounts for the greatest fraction of energy consumed for each manifold; 74% for the cast aluminum, 66% for the brazed aluminum tubular, and 61% for the nylon composite. Consequently, the manifold weight is the single most important determinant of life cycle energy. In the material production stage the energy difference between the two aluminum designs reflects the primary aluminum content associated with the tubular manifold. The material production energy for the tubular manifold is twice that of the cast manifold, even though the total mass of this manifold is only 56% of the mass of the cast manifold. End-of-life vehicle management processes have a very small energy requirement, indicated by a relatively low shredding energy of 0.097 MJ/kg. 3.1.2. Life cycle solid waste The life cycle solid waste profile of each manifold is indicated in Fig. 3, and contrasts sharply with the life cycle energy profiles where burdens are concentrated in the use phase. Fig. 3 shows that the multi-tube brazed manifold and the nylon composite generate the greatest
Life cycle energy of intake manifolds.
68
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
Fig. 3.
Life cycle solid waste of intake manifolds.
amount of life cycle solid waste. Both the nylon composite and the aluminum tubular manifold generate about twice as much life cycle solid waste, 4.2 kg and 3.9 kg, respectively, compared to the cast aluminum manifold, 2.2 kg. The majority of the solid waste associated with the aluminum tubular manifold is generated in material production of primary aluminum while a substantial amount of waste generation occurs in the end-of-life management stage for the nylon composite manifold. Material production of primary and secondary aluminum for a multi-tube brazed manifold results in 76% of its overall life cycle solid waste. As shown in Table 5, red mud, a caustic waste generated during alumina production, accounts for 87% of solid waste for primary aluminum processing. The major components of solid waste from a composite manifold in the material production stage, include mine tailings primarily from the copper used in brass inserts, combustion ash, mineral waste, sludge and polymer solids. On average, about 0.93 kg per kg of solid waste is generated from the production of materials for the composite manifold. Solid waste in the manufacturing stage is comprised of process waste from sand casting, product waste and energy waste. Sand casting waste consists of fume dust and an estimated 5% loss in recycling sand and salt slag. Product waste consists of an estimated 5% loss in recycling scrap generated from manifold production. The process and product wastes for a sand-cast manifold are 1.045 kg and 0.052 kg respectively. For a multi-tube brazed manifold, the process and product waste are 0.4 kg and 0.255 kg respectively. Process waste for a composite manifold in the manufacturing stage is primarily due to electricity generation and amounts to about 0.79 kg per intake manifold; product waste is 9.52 g per intake manifold. Solid waste during use primarily results from waste generated in the production of gasoline.
Retirement solid waste includes a 5% loss in recycling metals at the end-of-life of the vehicle. For the composite manifold, in addition to 5% metals waste, all the nylon (2.07 kg) ends up as solid waste. 3.1.3. Life cycle air pollutant emissions Fig. 4 shows life cycle pollutant emissions for the three manifold systems. The majority of pollutant emissions are in the form of nonmethane hydrocarbons (NMHC), NOx, CO and SO2. CO, NOx and NMHC releases are highest for a sand-cast manifold, followed by a multi-tube brazed manifold and a composite manifold. However, the trend is different for CH4, SO2 and PM-10. The contribution of the intake manifold to the total vehicle use phase emissions was estimated assuming that these emissions are proportional to gasoline consumption. Although this relationship is valid for carbon dioxide, this allocation is probably not accurate for the other pollutants that are controlled by the catalytic converter. 3.1.4. Life cycle greenhouse gas emissions Fig. 5 also illustrates that most greenhouse gases are released during the use phase for sand-cast and multitube brazed manifolds. For a sand-cast manifold, use phase greenhouse gas emissions represent about 76% of the life cycle total, while the use phase accounts for about 61% of total greenhouse gas emissions for a multitube brazed manifold. This difference is attributable to releases of CF4 and C2F6 during primary aluminum production. Although only 0.56 g of these fluorocarbons are released in the production of primary aluminum for a multi-tube brazed manifold, their global warming potentials are so much higher than CO2 (6300 for CF4 and 12,500 for C2F6 where CO2=1) that greenhouse gas emission in CO2 equivalents for producing multi-tube manifolds is 19.9 kg, compared to just 5.9 kg for sandcast manifolds. This, coupled with much lower use phase
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
Fig. 4.
Fig. 5.
69
Life cycle air pollutant emissions of intake manifolds.
Life cycle greenhouse gas emissions for intake manifolds in CO2 equivalents.
emissions (46.7 kg vs 83.8 kg for sand cast) due to the lighter weight of a multi-tube manifold, results in a significantly lower percentage of total greenhouse gas emissions occurring during the use phase. For similar reasons, the use phase accounts for only about 48% of total life cycle greenhouse gas emissions for a composite manifold; materials production accounts for 43%. Nitrous oxide (N2O, GWP=270) released during nylon production results in the highest greenhouse gas emissions of all the manifolds during this phase. N2O constitutes 71% of greenhouse emissions in nylon material production, CO2 for most of the remainder. In addition, the lighter weight of a nylon manifold results in the lowest CO2 emissions during use. Thus, greenhouse emissions are nearly evenly distributed between material production and use for a composite manifold rather than being concentrated in the use phase. It is apparent from this discussion that greenhouse gas emissions do not exactly parallel life cycle energy
requirements for these manifolds. Use phase energy for sand-cast, multi-tube brazed and composite manifolds accounts for 74%, 66% and 61% of life cycle energy respectively; greenhouse gas emissions for these manifolds, 76%, 61% and 48%. These differences result from the high global warming potential of halogenated carbons released during material production. 3.1.5. Life cycle water pollutant discharges Fig. 6 shows that the majority of water pollutants on a mass basis are in the form of dissolved solids, the highest of which are associated with a composite manifold, the lowest with a multi-tube brazed. 3.2. Life cycle cost analysis Table 8 shows that the life cycle costs of the two aluminum manifolds are similar. The life cycle cost of the composite manifold is approximately $10.76 more
70
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
Fig. 6.
Cumulative life cycle water pollutant discharges of intake manifolds.
Table 8 Life cycle costs of intake manifolds (in US dollars) Composite US Material cost Manufacturing costsa fixed variable Use phase costsb End of life costsc Salvage value Life cycle cost
German
Sand-cast US
German
Multi-tube brazed US German
$6.01
$6.01
$12.38
$12.38
$7.15
$7.15
$3.90 $50.16 $4.47 $0.42 $0.68 $58.27
$3.90 $50.16 $11.90 $0.42 $0.68 $65.70
$2.70 $38.66 $10.52 $1.81 $5.93 $47.76
$2.70 $38.66 $28.16 $1.81 $5.93 $65.40
$2.90 $40.80 $5.84 $1.00 $3.30 $47.24
$2.90 $40.80 $15.66 $1.00 $3.30 $57.06
a Manufacturing costs were estimated from data provided by Ford, for the German scenario analysis manufacturing costs were based on US conditions. b Use phase costs include both fuel and warranty costs. c End of life costs include transportation (dismantler, shredder, and landfill), disposal, and processing.
than that of the aluminum manifolds. The material cost of a sand-cast manifold is about $5.23 higher than that of multi-tube brazed and $6.87 higher than composite manifold. The higher material cost of a sand cast manifold is due to its higher weight compared to a multi-tube brazed manifold. The sum of the manufacturing and warranty cost for a multi-tube brazed manifold was estimated to be about $2.28 higher than a sand-cast manifold because of increased manufacturing complexity (extrusion, brazing and sand casting as opposed to sand casting only). The estimated manufacturing and warranty costs of a composite manifold is about $12.67 more than that of a sandcast manifold. Manufacturing accounts for the majority of life cycle costs for sand cast (87%), multi-tube brazed (93%) and composite (93%) manifolds. Ford’s manifold costs include both material purchase, manufacture, and warranty. Ford’s cost is estimated at $37.34 for a sandcast manifold, $41.44 for a multi-tube brazed manifold and $53.87 for a composite manifold.
Gasoline cost to the user of a sand-cast manifold, over a useful life of 150,000 miles, is about $4.62 more than that of a multi-tube brazed manifold, and $6.03 more than that of a composite manifold. These differences reflect the effect of weight on gas mileage. In the retirement stage, a sand cast manifold requires more to process, but has an aluminum scrap value that results in a net cost $1.82 lower than the multi-tube brazed manifold. A composite manifold requires the lowest processing cost in the retirement stage, because its major constituents are disposed to landfills rather than recovered. In contrast to the US scenario, the life cycle costs of the two aluminum manifolds, shown in Table 8, diverge greatly due to the higher cost of gasoline in Germany. This results in the heavier cast aluminum manifold having a life cycle cost $8.35 greater than the multi-tube brazed manifold and only a marginally lower cost than the composite manifold.
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
4. Discussion of system requirements The life cycle inventory analysis and life cycle cost analysis reveal significant tradeoffs among each of the three intake manifold designs. The selection of the preferred manifold design can be aided by evaluating these results with respect to system requirements. In addition to the environmental and cost requirements, performance requirements are also discussed. The intake manifold must accommodate vehicle system, powertrain subsystem, and engine specific requirements. Consequently, the manifold design should be evaluated in the context of these larger system boundaries. The decision analysis presented here, however, will be limited primarily to the manifold system. The decision making process is also influenced by the time horizon considered. Strategic planning can be an important element of the design process and lead to more ecologically sustainable design solutions. Decision makers may weigh greenhouse emissions more heavily when taking a long range perspective compared to a short term development cycle. 4.1. Environmental analysis The multi-criteria requirements matrices are a tool for identifying and organizing key requirements [3,10,41]. The set of internal and external environmental ‘requirements’ examined are presented in Table 9. Ford guidelines, corporate directives and policies as well as external requirements such as government regulations were identified. These environmental ‘requirements’ can be used to interpret results from the life cycle inventory analyses. Design decision making occurs in the context of the business and external forces impacting on the business and its products.
71
4.1.1. Energy Currently, no specific corporate guidelines or government regulations and policies encourage a reduction in the total life cycle energy. Several ‘requirements’ are directed at specific stages of the life cycle. For example, Ford’s Manufacturing and Environmental Leadership Program seeks to minimize facility energy consumption [42]. CAFE in the US and a voluntary pledge of the German auto industry to reduce CO2 emissions focuses on use phase energy. Corporate Average Fuel Economy (CAFE) is an important regulatory driver influencing vehicle fuel economy targets and weight targets. CAFE standards for passenger cars have been stagnant over the last decade [43] and new car corporate average fuel economy has followed a similar trend. Among the three manifold designs the nylon composite best meets all energy related requirements and consumes the least life cycle energy, as shown in Fig. 2. In this case, no tradeoffs emerge, although an impact assessment may consider the source of energy (coal, natural gas, petroleum, etc.). 4.1.2. Materials Ford internal requirements addressing life cycle materials include targets for recycled content of plastic resin, substance use restrictions, and vehicle/part weight reductions goals [42]. It is difficult to establish specific guidelines for interpreting the life cycle materials metrics presented in Table 10. Ideally, each design would minimize the total materials used including primary and secondary materials, maximize the total materials recycled, and reduce waste. It is well recognized that these criteria can easily conflict with other environmental objectives such as minimizing life cycle energy. The total material mass of the cast aluminum manifold design is greatest but the manifold utilizes secondary aluminum
Table 9 Internal and external environmental requirements Internal
External
Energy —Corporate citizenship —CAFE —Minimize facility energy (Manufacturing Environmental Leadership) —Voluntary pledge of German auto industry to reduce CO2 emissions —Meet platform fuel economy targets Materials —Ford targets for recycled content of plastic resin (D109, A120, —Reduce materials used, increase materials recycled, and reduce waste Manufacturing Environmental Leadership) —Substance use restrictions (WSS-M99P9999 also known as HEX9) —Reduce part/vehicle weight Waste —Protect health and environment (Policy Letter 17) —European guidelines for reducing waste going to landfill:maximum —Recyclability targets (Directive F-111) 15% by weight—2002 —Reduce manufacturing waste (A-120) maximum 5% by weight—2015 —Voluntary initiatives to reduce greenhouse emissions
72
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
Table 10 Materials metrics for intake manifolds (per IM basis) Cast Aluminum Product mass (kg) Primary material content Restricted substances (kg)
6.5 0.0% 0.017
and it is currently being recycled during end-of-life management of the vehicle. The brazed aluminum tubular manifold uses less total material but incorporates primary aluminum. This manifold is also recycled in the end-of-life phase. Both aluminum manifolds use phenol and formaldehyde to form molds for casting. The nylon manifold uses the least total material but incorporates the greatest quantity of primary material, which currently is not recycled during the end-of-life phase. 4.1.3. Waste The inventory category waste includes solid waste, air pollutant releases, and waterborne pollutant releases. The life cycle ‘waste’ inventory results were presented in Figs. 3–6. Interpretation of the inventory results presents several challenges, which are addressed and/or are being investigated as part of life cycle impact assessment methodology. The aluminum cast manifold generated the least amount of life cycle solid waste as was shown in Fig. 3. No guideline, however, exists at Ford, which seeks to minimize life cycle solid waste. Ford internal requirements address both manufacturing waste and end-of-life solid waste minimization. The minimization of material production solid waste is not specified as part of Ford’s material procurement guidelines. Consequently, Ford’s internal policy would favor the aluminum cast and the brazed aluminum tubular manifolds equally, even though a significant amount of solid waste, in the form of red mud, is generated with the brazed aluminum tubular system. The European guidelines for reducing the amount of waste going to landfill will probably lead to further emphasis on end-of-life waste compared with solid waste generated in other life cycle phases. 4.2. Cost analysis Life cycle costs can be grouped according to those direct costs incurred by Ford, and those costs occurring outside Ford’s cost domain. Ford manufacturing costs are a primary criterion for making a business decision. In addition, warranty costs are direct costs borne by Ford due to product defects or improper assembly of the manifold on the engine. The nylon composite manifold had the highest estimated manufacturing costs which were about $10 greater than the two aluminum manifold designs. The stainless steel EGR tube, only required for
Brazed Aluminum Tubular 3.62 25% 0.017
Nylon Composite 2.74 76% –
the nylon composite manifold, accounted for this differential cost. The life cycle costs incurred outside Ford’s domain during the use and retirement phases, represent costs to the customers and vehicle recyclers. The use phase gasoline cost to the customer over the lifetime of the vehicle was least for the composite manifold. The gasoline costs associated with the composite and the aluminum brazed tubular manifolds were about $6 and $5 less, respectively, than the cast aluminum manifold. The gasoline costs are much more significant in Germany and have a greater influence on vehicle purchasing decisions. Gasoline costs in the US are a relatively weak economic driver for reducing energy consumption in the use phase. The cost of end-of-life vehicle management may become an important criterion, since legislation is being discussed in Europe requiring OEM take back of automobiles at no cost. In this scenario, retirement costs would become part of the total Ford manifold cost. Fig. 7 indicates that, the aluminum cast and the brazed aluminum tubular manifolds would provide a greater end-oflife credit compared to the nylon composite manifold. Credits of $4.10 for the cast aluminum manifold and $2.30 for the brazed aluminum tubular manifold would accrue to the OEM. The salvage value of the brass and stainless steel associated with the nylon composite manifold offset waste disposal costs of glass-reinforced nylon. Otherwise, the nylon composite manifold would result in greater costs to Ford under the current European end-of-life management infrastructure. 4.3. Performance analysis Performance requirements are included to demonstrate design complexity. Each manifold alternative must meet basic performance criteria to become a viable candidate for a particular design application. A total of 20 performance requirements were used to evaluate each design alternative. As Table 11 shows, most of the requirements in the original Ford design analysis matrix were performance criteria. Weighting factors for the performance criteria are dependent on specific vehicle platform objectives. For example, the NVH-Acoustical (noise, vibration, and harshness) criterion would be weighted higher for a luxury car relative to an economy car. The cast aluminum manifold generally had higher rankings compared with the nylon composite or the brazed alumi-
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
Fig. 7.
73
Life cycle costs for intake manifolds for US and Germany.
Table 11 Original Ford requirements matrix
Requirements
Ranking Cast Aluminum
120k Durability First Time Quality Capable Airflow/Performance Weight Fastener Compatibility Joint Sealing Material Dimensional Stability Flammability Resistance High Temperature Performance Low Temperature Performance Positive Pressure Capability NVH-Structural NVH-Acoustical Prototype Lead Times Prototype Tooling Cost Production Lead Times Variable Cost Production Tooling Cost Appearance Established Supply Base Manufacturing Flexibility Component Integration Opportunity Design Flexibility
num tubular manifolds. The nylon composite manifold, however, was preferred for several important criteria, including, first time quality capability, weight, and component integration opportunity. The three manifolds investigated in this project meet these criteria.
10 6 6 4 10 8 10 10 10 10 10 10 8 8 8 8 8 8 6 10 6 4 8
Brazed Aluminum Tubular 8 4 8 6 6 8 6 10 8 10 8 6 4 6 6 6 8 4 6 4 4 2 6
Nylon Composite
8 10 8 10 2 6 4 2 2 2 4 4 2 4 2 4 6 2 8 6 2 8 6
4.3.1. Manufacturability 4.3.1.1. Composite manifold It can be seen from Table 11 that the composite manifold involves three different materials and requires eleven unit processes for manufacturing. The lost core process consists of five dif-
74
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
ferent unit processes. The cycle time for injection molding is 1.5 min, the cycle time for core casting is 3 min and the cycle time for core melt out is 45 min. In the lost core process, maintaining an appropriate core casting temperature and controlling core dimensional change during injection molding presents significant challenges to manufacturers. Getting the time-temperature cycle right on the core casting tool is critical [44]. If cores are cooled too fast they crystallize and become brittle, but if cores are cooled too slowly portions can still remain molten when the core is overmolded by nylon resin. The most critical part of the lost core process is accounting for melt loss of cores during injection molding. Since the tin-bismuth core alloy has a lower melting temperature (320°F–160°C) than nylon resin (491°F–255°C), some core metal may get melted when it is overmolded with molten nylon resin. Nylon resin loses its heat while melting part of the core layer and also undergoes stress relief and shrinkage during the melt-out stage. Therefore, in lost core process design, these dimensional changes are built into the tool design [44]. Because the core material has to be melted every time, lost core molding is a very energy intensive process. In addition, the stainless steel EGR tube increases overall complexity because it requires three different manufacturing processes. 4.3.1.2. Sand-cast manifold The sand-cast manifold was the only one-piece manifold studied. As indicated in Table 6, sand casting involves five different unit processes. A typical cycle time for manufacturing a sandcast manifold is 14 min. This includes 1 min for core fabrication, 2 min for casting, 5 min for cooling, 0.5 min for pre-machining pressure testing, 0.5 min for machining and 2 min for washing, assembly, testing and packaging. The tool life for a typical aluminum manifold is about 250,000 cycles. The die life is about 1×105 to 2×105 mold parts before reconditioning. 4.3.1.3. Multi-tube brazed manifold The multi-tube brazed Escort manifold is comprised of a cast aluminum flange, four bent aluminum tubes, and an air collection chamber joined together by brazing. The aluminum tubes and the collection chamber are manufactured by extrusion. After extrusion, aluminum tubes are bent into desired shapes by a movable mandrel. The casting is placed into a die and pressurized hydraulic fluid turns out the four openings from inside [30]. Table 6 indicates that the multi-tube brazed manifold involves nine different manufacturing unit processes that include five processes for sand casting. The cycle time for a multi-tube brazed manifold was not available, but it is expected to be higher than that of a sand-cast manifold because of extrusion and brazing. Brazing poses a significant challenge to the manufacturing, performance and quality of
the multi-tube brazed manifold. First it increases the cycle time because each joint has to be brazed separately and then tested for leaks. Secondly, brazing, if done incorrectly could lead to leaks and quality problems. 4.3.2. Use The smoother wall of the multi-tube brazed manifold is expected to lead to less friction loss compared to the rough-walled, sand-cast manifold. This theoretically translates into higher volumetric efficiency and higher power output at the same throttle opening. However, Ford test engineers reported no significant difference in power between engines equipped with rough-walled, sand-cast manifolds and smooth-walled, composite manifolds at part throttle. At full throttle a 2% increase in power for the composite manifold was obtained. Similar conclusions can be inferred about smoother-walled, multi-tube brazed manifolds. Ford’s manifold design group reported that composite manifolds deform to the shape of the engine where they are used and therefore cannot be remounted on another vehicle after retirement. Ford’s manifold engineers could not confirm reports of defects due to heat deformation for the 1995 Countour manifold. The stainless steel EGR tube is expected to transfer most of the heat away from the manifold. Seven warranty claims related to composite manifolds were filed for a seven-month period during which 55,000 1995 Contours were sold. This is a defect rate of 0.13 per 1000 vehicles. Because the sand-cast manifold was not used in actual vehicle production, warranty data for this manifold are not available. For the multi-tube brazed Escort manifold, 262 warranty claims were filed in the last five years during which 1,438,593 vehicles were sold. This is a defect rate of 0.18 per 1000 vehicles. These warranty data include manufacturing flaws, assembly errors, mis-bins (wrong parts serviced) and accident repairs.
5. Conclusions This demonstration project with Ford applied the life cycle design framework to air intake manifold design. This project was successful in providing environmental, cost, performance, regulatory, and policy data for enhancing the design analysis of three alternative air intake manifolds: cast aluminum, brazed aluminum tubular, and nylon composite. Significant tradeoffs among the three designs were highlighted and the value of the life cycle design framework was discussed. The multicriteria requirements’ matrix was useful in identifying and organizing key requirements for design analysis. The life cycle inventory analysis indicated significant environmental tradeoffs among alternative manifold designs. The life cycle energy consumption for the cast
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
aluminum, brazed aluminum tubular, and nylon composite manifolds were 1798 MJ, 1131 MJ, and 928 MJ per manifold, respectively. The cast aluminum manifold, however, generated the least life cycle solid waste, 218 kg per manifold, whereas the brazed aluminum tubular and nylon composite manifolds generated comparable quantities of 418 kg and 391 kg, respectively. These life cycle inventory analysis results were interpreted with respect to Ford internal environmental policies, guidelines, and goals as well as external environmental requirements such as existing and proposed government policies and regulations. No specific Ford policy, however, states that the total life cycle environmental burdens for each automotive part and component should be minimized. Rather different policies and guidelines address discrete stages of the life cycle. The same holds true for life cycle costs including manufacturing costs borne by Ford, customer costs, and end-oflife management costs. Life cycle optimization of environmental and cost objectives is therefore not necessarily an expected outcome. This project revealed several organizational factors affecting the successful implementation of life cycle design projects. Comprehensive evaluation of the total life cycle system necessitated the participation of a cross functional team with a broad range of expertise. This project educated many members of the team on the life cycle design methodology. The multiobjective analysis served to introduce the project team to the full spectrum of issues constraining the manifold system. It was recognized, due to the model complexity and data intensity, that a comprehensive evaluation should not be performed in the final stages of design but rather it would be performed in the planning stages. As a planning tool for product development life cycle design can highlight opportunities for improvement by identifying major environmental burdens, costs, regulatory and policy issues to target. As a planning tool, alternative materials and design strategies can be explored. The project team also discussed the challenge of predicting trends in future end-of-life management infrastructure that could impact new vehicles retired ten to twenty years in the future. This time lag introduced a significant level of uncertainty into the design analysis process. Several members of the project team advocated characterizing the different environmental burdens into a single score to facilitate the use of the life cycle assessment methodology by design engineers. A variety of techniques were investigated, including translating the environmental burdens into monetary costs, applying the critical volume approach, the environmental theme method, and other impact assessment methods [45]. None of the approaches was found to be acceptable to the project team, due to limitations in evaluating parameters needed for these different models. A single score approach may also limit the design team from exploring
75
how major environmental burdens are distributed across the product life cycle. In addition, the direct relationship between these burdens, and cost, performance, regulatory and policy factors can be more clearly understood if burdens are itemized. Emphasis of the project team was more on framing the design decision rather than on actually recommending a preferred manifold design. Each manifold design had a superior set of attributes. The project team favored the aluminum brazed tubular and nylon composite manifolds over the sand cast aluminum manifold, due to their weight differentials. For this manifold application, the aluminum tubular design offered significant manufacturing cost savings relative to the nylon composite design. This may have overshadowed the slightly better life cycle energy performance of the nylon composite manifold. An important benefit of the life cycle design framework is that it clarifies the complex set of factors that influence the likelihood for success of a business decision. Tradeoffs are made explicit, and interrelationships between design objectives are made apparent. It should be recognized that this paper presents results obtained during the demonstration project, which may or may not be representative of current conditions. For example, nylon manifolds can be manufactured by injection molding and vibrational welding [14]. An air intake manifold is only one component of the powertrain system, which is part of the total vehicle system. Consequently, it makes only a relatively small contribution to the overall environmental burdens of an automobile. More widespread application of the life cycle design methodology to other vehicle components and systems, however, can result in substantial opportunities for improvement. This project served to demonstrate the value of life cycle systems thinking in design, and will hopefully be extended to other parts and components, as well as higher level vehicle systems in the future.
Acknowledgements This research was funded by the United States Environmental Protection Agency (EPA) under Cooperative Agreement number CR822998-01-0 to the University of Michigan. Mention of trade names or commercial products does not constitute endorsement or recommendation for use by US EPA. We thank the following members of the Ford Project Team for collaborating with us on this project: Powertrain Engineering (Wayne Koppe, Fred Heiby, Cymel Clavon, David Florkey, and Mitch Baghdoian) Scientific Research Laboratory (John Sullivan, Mia Costic), Casting Operations (George Good), Advanced Vehicle Technology (Mike Johnson) and Environmental and Safety Engineering (Philip Lawrence, Bernd Gottselig). We also thank Ken Martchek
76
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
from Alcoa and William Haaf and David Doyen from DuPont for providing material production and manfuacturing inventory data. [18]
References [19] [1] Keoleian GA, Menerey D. Life cycle design guidance manual: Environmental requirements and the product system. Cincinnati, OH: US EPA, Office of Research and Development, Risk Reduction Engineering Laboratory. 1993. EPA/600/R–92/226. [2] Keoleian GA, Menerey D. Sustainable development by design: review of life cycle design and related approaches. J of Air and Waste Management Assoc 1994;44(5):645–68. [3] Keoleian GA, Koch J, Menerey D. Life cycle design framework and demonstration projects: Profiles of AT&T and Allied Signal. Cincinnati, OH: US Environmental Protection Agency, National Risk Management Research Laboratory. 1995. EPA/600/R95/107. [4] Alexander B, Barton G, Petrie J, Romagnoli J. Process synthesis and optimisation tools for environmental design: methodology and structure. Computers and Chem Eng 2000;24:1195–200. [5] Azapagic A. Life cycle assessment and its application to process selection, design and optimisation. Chem Eng J 1999;73:1–21. [6] Azapagic A, Clift R. Life cycle assessment and multiobjective optimisation. J of Cleaner Prod 1999;7:135–43. [7] Kainz RJ, Simpson M, Moeser WC. Life cycle management: A solution for decision making in the global market. In: SAE International Congress & Exposition. Warrendale, PA: Society of Automotive Engineers. 1994. Technical Paper 940575. [8] Keoleian GA, Spatari, S, Beal RT. Life cycle design of a fuel tank system. US Environmental Protection Agency, Office of Research and Development. Cincinnati, OH: National Risk Management Research Laboratory. 1997. EPA 600/R-97/118. [9] Keoleian GA, Spatari S, Beal RT, Stephens RD, Williams RL. Application of life cycle inventory analysis to fuel tank system design. Internat J of Life Cycle Assessment 1998;3(1):18–28. [10] Keoleian GA, McDaniel JS. Life cycle design of instrument panels: A common sense approach. In: SAE International Congress and Exposition. Warrendale, PA: SAE International. 1997. Technical Paper 970695. [11] Keoleian GA. Is environmental improvement in automotive component design highly constrained? J of Ind Ecology 1998;2(2):103–18. [12] Kar K, Keoleian GA. Application of life cycle design to aluminum intake manifolds. I: SAE International Congress and Exposition. Warrendale, PA: Society of Automotive Engineers. 1996. [13] Keoleian GA, Kar, K. Life cycle design of air intake manifolds: Phase I: 2.0 L Ford Contour air intake manifold. National Risk Management Research Laboratory, Office of Research and Development, US Environmental Protection Agency. EPA 600/R99/023. [14] Spitzley D, Keoleian GA. Life cycle design of air intake manifolds: Phase II: Lower plenum of the 5.4 L F-250 air intake manifold, including recycling scenarios. Cincinnati, OH: Office of Research and Development, National Risk Management Research Laboratory, US Environmental Protection Agency. EPA/600/R01/059. [15] Kar K, Keoleian GA, Spitzley DV, Malone K, Whitney S. Life cycle design of in-mold surfacing film. Cincinnati, OH: US EPA National Risk Management Research Laboratory. EPA/600/R01/058. [16] Reppe P, Keoleian G, Messick R, Costic M. Life cycle assessment of a transmission case: Magnesium vs aluminum. 1998. SAE International, Paper No. 980470. [17] Shen D, Phipps A, Keoleian G, Messick R. Life-cycle assessment
[20]
[21]
[22]
[23]
[24]
[25] [26] [27] [28]
[29] [30]
[31]
[32] [33]
[34]
[35]
[36] [37] [38]
of a powertrain structural component: Diecast aluminum vs hypothetical thixomolded magnesium. In: SAE International Conference and Exposition. Detroit, MI: SAE International. 1999. Paper No. 1999-01-0016. Keoleian GA, Spitzley D. Guidance for improving life-cycle design and management of milk packaging. J of Ind Ecology 1999;3(1):111–26. Spitzley D, Keoleian GA, McDaniel J, Menery D. Life cycle design of milk and juice packaging. Cincinnati, OH: US Environmental Protection Agency, Office of Research and Development, National Risk Management Reserach Laboratory. 1997. EPA/600/R-97/082. Lewis G, Keoleian GA. Life cycle design of amorphous silicon photovoltaic modules. Cincinnati, OH: US Environmental Protection Agency, Office of Research and Development, National Risk Management Research Laboratory. 1997. EPA/600/R-97/081. Keoleian GA, Lewis GM. Application of life-cycle energy analysis to photovoltaic module fesign. Progress in Photovoltaics 1997;5:287–300. Vigon BW, Tolle DA, Cornary BW, Latham HC, Harrison CL, Bouguski TL, et al. Life cycle assessment: Inventory guidelines and principles. Cincinnati, OH: US Environmental Protection Agency, Risk Reduction Engineering Laboratory. 1993. EPA/600/R-92/245. SETAC. Guidelines for life-cycle assessment: A code of practice. In: Workshop Report—Guidelines for Life-Cycle Assessment: A Code of Practice. Pensacola, FL: Society of Environmental Toxicology and Chemistry. 1993. ISO. Environmental management—life cycle assessment—principles and framework. Geneva: International Organization for Standardization. 1997. Bustamalti T. Personal communication. Deeco Industries. 1996 (January). Steel Recycling Institute. Recycling scrapped automobiles. Pittsburgh, PA: SRI. 1994. Martchek K. Personal communication. Alcoa. 1995 (August). Doyen D, Haaf B. Material production inventory data for glassreinforced nylon (PA6.6 GF33) and Brass (UNS 36000). Wilmington, DE: Dupont. 1995. Franklin Associates. Material production database. Prairie Village, KS. 1995. Eyerer P, Schuckert M, Dekorsy T. Life cycle analysis of automotive air intake manifolds of aluminum and PA 6.6 GF 3. Stuttgart, Germany: Institut fu¨ r Kunststoffpru¨ fung und Kunststoffkunde, Universita¨ t Stuttgart. 1992. Aluminum Company of America Alcoa. Environmental life cycle considerations of aluminum in automotive structures. Southfield, MI: Automotive Structures. 1994. Environmental Papers 1–4. International Iron and Steel Institute. Competition Between steel and aluminum for the passenger car. Brussels. 1994. Sullivan JL, Hu J. Life cycle energy analysis for automobiles. In: SAE Total Life Cycle Conference. Warrendale, PA: Society of Automotive Engineers. 1995. SAE Paper 951829. Harnisch J, Borchers, R, Fabian P. Estimation of tropospheric trends (1980–1995) for CF4 and C2F6 from stratospheric data. In: Pollution of the Troposphere and Stratosphere. 1995. US Environmental Protection Agency. The lean and green supply chain: A practical guide for material managers and supply chain managers to reduce costs and improve environmental performance. 2000. p. 58. Energy Information Agency. International Energy Annual. US: US Department of Energy, 1993 p 94–95. Economics of Recovery and Recycling. American Plastics Council. 1994. McKinley MD, Jefcoat IA, Herz WJF. Air emissions from foundries: A current survey of literature, suppliers and foundrymen. AFS Transactions 1993;105:979–90.
G.A. Keoleian, K. Kar / Journal of Cleaner Production 11 (2003) 61–77
[39] American Metals Market. July 19, 1996. p. 6. [40] Osterberg D. Personal communication. Huron Valley Steel. 1995. [41] Keoleian GA. Pollution prevention through life cycle design. In: Freeman HM, editor. Pollution Prevention Handbook. New York: McGraw-Hill; 1995. p. 253–92. [42] Ford Information. Detroit, MI: Ford Motor Company. 1996. [43] Keoleian GA, Kar K, Manion M, Bulkley J. Industrial Ecology
77
of the Automobile: A Life Cycle Perspective. Warrendale, PA: Society of Automotive Engineers, 1997. [44] Naitove MH. Blow molding. Plastics Tech 1994;August:45–7. [45] Baumann H, Rydberg T. Life cycle assessment: A comparison of three methods for impact analysis and evaluation. J of Cleaner Prod 1994;2(1):13–20.
