Interest in assessing the sustainability performance of manufacturing processes and systems during product design is increasing. Prior work has investigated approaches for quantifying and reducing impacts across the product life cycle. Energy consumption and carbon footprint are frequently adopted and investigated environmental performance metrics. However, challenges persist in concurrent consideration of environmental, economic, and social impacts resulting from manufacturing processes and supply chain networks. Companies are striving to manage their manufacturing networks to improve environmental and social performance, in addition to economic performance. In particular, social responsibility has gained visibility as a conduit to competitive advantage. Thus, a framework is presented for improving environmental and social performance through simultaneous consideration of manufacturing processes and supply chain activities. The framework builds upon the unit manufacturing process modeling method and is demonstrated for production of bicycle pedal components. For the case examined, it is found that unit manufacturing processes account for 63–97% of supply chain carbon footprint when air freight transport is not used. When air freight transport is used for heavier components, transportation-related energy consumption accounts for 78–90% of supply chain carbon footprint. Similarly, from a social responsibility perspective, transportation-related activities account for 73–99% of supply chain injuries/illnesses, and days away from work when air freight transport is used. Manufacturing activities dominate the impacts on worker health when air freight transport is not used, leading to 59–99% of supply chain injuries/illnesses, and days away from work. These results reiterate that simultaneous consideration of environmental and social impacts of manufacturing and supply chain activities is needed to inform decision making in sustainable product manufacturing.

## Introduction

High energy demands have led to the increasing consumption of nonrenewable, fossil-based energy resources [1]. The combustion of fossil fuels to generate energy is accompanied with the emission of greenhouse gases (GHGs), often measured in terms of carbon dioxide equivalent (CO2 eq.), or carbon footprint. Climate change is one major concern of GHGs and has been identified as one of the most complex sustainability challenges faced by decision makers globally [2]. Potential risks include coastal flooding due to increased levels of seawater and ecosystem damage due to increases in atmospheric temperature [3]. Because GHGs are major emissions from industrial and transportation operations, assessment methods and frameworks are needed to evaluate direct and indirect emissions from manufacturing and supply chain activities during product design, process planning, and supplier selection activities. Different supply chain configurations for a product can introduce a variety of unit manufacturing processes and related machine capabilities, as well as variation in energy use, carbon footprint, and worker health impacts. These factors all impact the sustainability performance of a product.

Although the importance of sustainable development has increased, many challenges exist in developed and developing countries [4]. While some manufacturers, service firms, and logistics companies are taking serious steps toward integrating sustainability principles into their policies and practices, environmental and social performance measures are often not accounted for in product design and manufacturing decisions. The presented work specifically aims at examining the environmental and social aspects of sustainability through simultaneous consideration of several impact factors. Since economic analysis is typically conducted as a part of engineering decision making and is a common measure of success, it is not demonstrated herein for brevity. However, traditional economic assessment approaches can be readily integrated into the presented framework. To reduce energy consumption, associated carbon footprint, and worker health impacts of product manufacturing, simultaneous consideration of manufacturing processes and supply chain activities needs to be undertaken from a cradle-to-gate life cycle perspective [5]. The consequences of social impacts have received little attention in engineering and business decisions compared to economic and environmental issues [6,7]. Two indicators of social performance related to worker health, nonfatal occupational injuries and illnesses and days away from work, are readily measurable metrics and are in focus here. Other social metrics have been suggested (e.g., investment in the community and workforce demographics) [8,9]. Such metrics are supplier-specific and, thus, more relevant to supplier management activities than early product and process design decisions considered here.

The research presented herein serves two main objectives. First, it provides a brief review of the literature between 1990 and 2015 to summarize past and current findings and methods related to the environmental and social aspects of sustainable products, processes, and systems. The review helps to identify research gaps and support the novelty of the work. Second, it proposes a method, complete with a decision framework, to support decision making for manufacturing and supply chain sustainability using parametric models to quantify a select set of performance metrics. Thus, engineers and other decision makers can evaluate the effect of choosing different supplier locations based on the defined manufacturing processes and transportation modes. An application of the method is presented to illustrate the framework and potential of the approach in enhancing the decision making in sustainable product design and manufacturing by using the following environmental and social metrics: energy use, carbon footprint, nonfatal occupational injuries and illnesses, and days away from work.

### Social Responsibility.

Companies across the world have come to understand that business success is dependent upon protection of the environment and the welfare of current and future generations [10]. Recent work has explored socially responsible design and manufacturing. Jorgensen et al. reviewed social life cycle assessment (SLCA) to provide a picture of the present landscape and to highlight methodological differences and general shortcomings [11]. Parris and Kates reviewed 12 initiatives defining social indicators, which identified indicator sets ranging in size from 6 to 255 and in scope from local to global [12]. Golini et al. investigated the effect of site competence (defined as the number of operations a facility holds responsibility for) on a facility's economic and social performance [13]. Using a survey-based approach, they found a high correlation, indicating that more capable facilities also have better sustainability performance. Yawar and Seuring provided a conceptual framework and insights into the management of social issues by linking supply chain actions and performance outcomes. Social issues included labor conditions, human rights, health and safety, and inclusion of disabled/marginalized people, among others [14]. Hutchins and Sutherland provided an input–output modeling approach to consider social sustainability in supply chain decision making [15]. Indicators included labor equity, healthcare, injuries, and philanthropy.

Related work developed a sustainability assessment methodology that integrates unit process modeling and life cycle inventory techniques for improving the accuracy of product economic, environmental, and social impact assessment [16]. Wilhelm et al. examined triple bottom line sustainability in the mobile phone industry using a life cycle approach [17]. A two-part framework was used to consider improvement opportunities that influence social impacts and the related economic and environmental impacts. Zhang and Haapala [18] developed an approach to assess sustainability impacts at the work-cell level through economic assessment, environmental assessment, and social assessment. Results are combined using a subjective weighting method based on pairwise comparison and an outranking algorithm.

Gunasekaran and Spalanzani [10] developed a framework for sustainable business development (SBD) that suggests future research directions and includes tools, techniques, and performance measures/metrics. The framework was based on seven building blocks: sustainable challenges and problems, advances in SBD in marketing and sales, sustainable product and process design, and sustainability considerations in supply operations, production operations, distribution operations, and through reverse logistics. Along with performance measures for economic and environmental aspects such as net profit, return on investment, carbon footprint, and human natural resource consumption, Székely and Knirsch [19] identified social metrics for human rights, labor issues, supplier relationships, community initiatives, and corporate philanthropy.

Ma and Okudan Kremer [20] considered various aspects of sustainability and subjective and vague perceptions of designers to develop a systematic way for determining end-of-life (EOL) strategies for retired product components. Job creation, human toxic potential, and land use were used to assess social impacts. Chen et al. [21] applied a societal viewpoint for analyzing direct digital manufacturing for various traditional manufacturing paradigms (i.e., craft production, mass production, and mass customization). Sancha et al. [22] developed a conceptual model to analyze the impact of assessment and collaboration on the social performance of the buying firm and supplier and to investigate the relative effectiveness of these two practices.

The limitations of prior work described in Secs. 1.21.4 and the following subsections are summarized in Sec. 1.5.

### Environmental Responsibility.

Environmental responsibility has emerged as a critical aspect of production decisions since the 1960s due to industrial growth and expansion, particularly in developed countries [23]. Researchers and practitioners across many disciplines in academia and industry continue to vigorously pursue the integration of environmental responsibility principles into design of processes, products, and systems to meet market and governmental expectations for sustaining quality of life for humans [2427]. Systematic approaches have been developed to meet customer expectations and address industrial commitment to environmental policies and legislation for sustainable product design.

Improving business environmental performance has been widely addressed through environmentally benign manufacturing (EBM). This system-level approach adopts green manufacturing processes, which are designed to be material and energy efficient [28]. Duflou et al. [29] provided a systematic assessment of energy and resource efficiency methods and techniques in the domain of discrete part manufacturing. A review of analytical tools and frameworks aiding environmental competitiveness was conducted by a World Technology (WTEC) panel more than a decade ago [30]. Approaches were discussed for evaluating impacts and possible improvements from an environmental perspective, including metrics and data-based approaches, design for environment, and life cycle assessment (LCA). Understanding the characteristics, inputs, and outputs of products, processes, and systems is imperative in defining, assessing, and measuring the related environmental impacts [26]. Design for environment, also known as ecodesign, focuses on the environmental, health, and safety issues related to a product's complete life cycle, beginning with material extraction and ending with disposal or end-of-life [31]. It has been hypothesized that product environmental concerns can be designed-in during the early design phase to alleviate impacts [32].

Fullan [33] posited that sustainability requires coordination of the challenges and obstacles present in a system due to continuous improvement and human values for quality development. Indeed, this encompasses many interpretations and warrants discussion. Hargreaves and Fink [34] argued that sustainability is not necessarily the act of sustaining a particular resource; rather, it is the development of processes and systems in a way that guards against sacrificing natural resources in the present and the future. The U.S. Environmental Protection Agency (EPA) defines sustainability as protecting the environment and human health from the exhaustive consumption of the ecological resources [35]. While there is disagreement on how sustainability is attained, definitions affirm the need for practices that responsibly utilize current resources and are cognizant of the needs of future generations. LCA is the most popular and widely used method to comprehensively assess environmental impacts [36] and is supported by a set of standards through initiatives led by the United Nations (UN) and the International Organization for Standardization, e.g., development of the ISO 14000 series in the 1990s [37]. LCA is fundamentally process-based and much work over the past two decades has sought to support manufacturing process-level analysis.

### Sustainable Manufacturing Process Research.

The need for manufacturing process modeling and development was stressed by the U.S. National Research Council (NRC) two decades ago [17]. Further, in 2001, a global study of manufacturers concluded that models are needed to understand economic value and environmental performance of product design and manufacturing [30]. A number of studies have been undertaken to meet these objectives over the ensuing years. Giachetti [38] developed an analytical tool to support design for manufacturing (DFM), including the development of a database that stores material characteristics. Lin and Polenske [39] applied input–output process modeling to investigate the influence of different steelmaking processes on the cost of disposed wastes. Similarly, Sutherland and Gunter [24] proposed a general input–output process-based methodology to assess the effect of process planning on waste generation.

Hernandez-Matias et al. [40] found that most work in manufacturing process modeling was either theoretically driven or centered on a particular process. The study revealed that the most widely investigated metrics include economic value of production, process cycle time or throughput, and optimization of labor. Rajemi et al. [41] developed a methodology to model the effect of machining process parameters on energy consumption, which was illustrated for turning. An ISO standard has emerged to define the environmental performance of machine tools during the use phase [42]. Diaz et al. [43] investigated the impact of different material removal rates on the specific energy consumption (energy per volume of material removed) of a milling machine and examined the tradeoff between power demand and machining time to confirm that the total energy consumption is not increased because of increased loads due to faster material removal. Gutowski et al. [44] investigated the energy requirements of manufacturing processes and concluded that specific energy consumption varies greatly for various unit processes. Haapala et al. [45] developed models to estimate the materials and energy use and wastes for processes involved in the production of large steel products (e.g., bending and welding). Linke et al. provided efficiency indicators for finishing processes, which are computed as changes in process performance or part quality divided by the necessary resources [46]. They applied three efficiency indicators, based on average roughness, average peak-to-valley height, and subjective part quality, to a case study on vibratory grinding. Recent efforts have sought to unify these various approaches, as discussed below.

Madan et al. [47] developed a science-based reference for estimating the energy consumption of unit manufacturing processes with three objectives: prediction, benchmarking, and improvement. Further, the unit process life cycle inventory (UPLCI) effort, in the U.S. and the CO2PE! Initiative, in Europe, has established a method for manufacturing process analysis to support LCA [48,49]. The method quantifies energy use for manufacturing processes using data from literature or experimentation and has been demonstrated for drilling [50] and other unit processes [25]. Product environmental performance assessment often focuses on reducing energy consumption and related carbon footprint of manufacturing. A recent study by Branker et al. [51] developed a model based on microeconomics that linked the reduction of energy consumption and associated carbon footprint to the reduction of total product cost. Nava [52] investigated a method to minimize energy consumption and related carbon footprint in metal forming by estimating the energy of the mechanical work, converting energy to CO2, and providing an optimization approach to minimize the resultant CO2. Other researchers have developed models to correlate electrical energy use to the GHG emissions for traditional manufacturing processes [53,54].

Fang et al. [55] investigated the application of linear programing to optimize the peak power load and energy use of manufacturing systems. Dietmair and Verl [56] presented a model for forecasting machine energy consumption, which was then illustrated for milling machines. In order to identify optimal energy control actions for typical manufacturing systems, Sun and Li [57] used the Markov decision process (MDP) and established an analytical model to evaluate the effect of reduced power demand of typical manufacturing systems under demand response. Sun et al. [58] analyzed the effect of energy savings opportunities from the system level and proposed an optimal state finding and updating approach to enhance energy efficiency for sustainable production systems. In addition to process-level and facility-level sustainability performance evaluation, recent work has examined supply chain sustainability, as discussed in Sec. 1.4.

### Sustainable Supply Chain Research.

Ilgin and Gupta [59] presented a comprehensive review of current advances and activities in environmentally conscious manufacturing and product recovery (ECMPRO) consisting of 540 peer-reviewed publications from the 1960s to 2010. The review focused on trends related to four phases of the product life cycle: environmentally sustainable product design, supply chain design, remanufacturing, and disassembly activities. Seuring and Müller [60] synthesized an extensive review of 191 scholarly papers on sustainable supply chain management from 1994 to 2004. Their review offered evidence that studies were deficient in connecting energy use and GHG emissions during supply chain operations, which was reiterated by Cholette and Venkat [61]. The sustainability of a supply chain network is subject to many factors, e.g., technology use, energy use, network density and route, inventory policy, trade policy, and shipment policy [62]. Saunders and Barber [63] found that GHG emissions related to various transportation methods, loads, and routes for raising lambs account for a 34% emission increase in the UK compared to shipping the same product from New Zealand to the UK. Ibbotson and Kara [64] examined a dozen supply chain network designs differing in transportation modes and routes, loads to be transported, and energy profile per location. Their study concluded that rail transport exhibited a 3–9% lower carbon footprint than other modes. Further, it has been reported that using local suppliers can cut the carbon footprint of rail transport by 10% [65].

Chiu et al. [32] combined a graph theory approach with LCA for generating product design concepts to account for cost and carbon footprint. Their approach was applied to study a global bicycle supply chain and showed potential in informing decision makers at the design stage. Arnette et al. provided a comprehensive review in design for X (DFX) literature and techniques (X refers to an activity, feature, or goal). They also investigated the future directions of the field and created a design for sustainability taxonomy that integrates the current DFX techniques in the categories [66]. Sundarakani et al. [62] developed a quantitative, heat flux modeling approach to examine the carbon footprint of a supply chain network, by considering stationary (manufacturing) and nonstationary (transportation) emissions sources. They applied Eulerian and Lagrangian transport models to calculate the carbon footprint due to a range of gaseous emissions. This approach enabled visualization of emissions for each node of an automotive supply chain network to facilitate decision making. Bevilacqua et al. [67] studied the effect of transportation type, combinations of transportation type and route, and selection of suppliers to minimize environmental impacts for a textile company.

In addition to the research conducted on sustainable manufacturing and supply chain designs, environmental assessment tools and methods have been developed. Seuring [7] reviewed papers on quantitative models for forward supply chains and provided future research directions. He determined that the complexity of modeling social aspects might inhibit it more frequently being taken into account during decision making. Govindan et al. [68] provided a conceptual model to assess the impact of lean, resilient, and green supply chain management practices on social, economic, and environmental aspects. The conceptual relationships identified can enable development of new theoretical approaches and empirical research in supply chain management and sustainability. Ahi and Searcy [69] identified a number of organizational or operating environment specific factors from the literature to investigate how these factors affect sustainability across the supply chain. They found that factors such as risk management, customer demands, collaboration with suppliers, and regulations are fundamentally context dependent. Garcia and You [70] focused on supply chain design and its integration with operations in the supply chain. Besides providing an overview of opportunities, such as increasing importance of energy and sustainability, the authors addressed relevant technical challenges in supply chain design, categorizing them as multiscale challenges, multi-objective and sustainability challenges, and multiplayer challenges.

### Limitations of Prior Work.

Frameworks and tools mentioned above are among the most widely used for assessing the sustainability performance of products, processes, and systems, yet several limitations have been identified as follows [21,7175]:

• time intensity and error proneness in data collection

• inaccuracy in defining the scope and system boundaries

• variability of the process model parameters

• uncertainty of chemical and material properties

• poor accommodation of predefined processes and systems

• poor fidelity in modeling impacts of manufacturing processes and equipment

• little to no quantitative analysis of worker health and safety impacts of manufacturing

• no consideration of impacts of design and manufacturing changes communities and work environments

From the foregoing literature review, two additional primary limitations have been identified. First, LCA tools provide an overall assessment of the environmental performance of products, processes, and systems, yet the presence of the above identified deficiencies can yield inaccurate and uncertain conclusions. The development of parametric unit manufacturing process models can overcome several of these deficiencies and yield more accurate results and definitive conclusions. In particular, identification and manipulation of impactful process parameters would become accessible during process planning. Second, simultaneous consideration of manufacturing processes and supply chain design alternatives requires ad hoc approaches to combine process- and system-level analyses. Thus, research is needed to develop and integrate bottom-up (process-level) and top-down (system-level) models to facilitate multi-objective cradle-to-gate sustainability assessment.

To address these limitations, a methodology and framework are presented herein for modeling manufacturing and transportation processes to characterize the sustainability performance of product, process, and supply chain designs in an integrated manner. The approach links product data to multiple sustainability metrics across supply chain operations, including manufacturing and transportation processes. The framework utilizes a process-based parametric modeling approach to simultaneously evaluate unit manufacturing processes and supply chain network scenarios, and is amenable to a range of processes and sustainability metrics (e.g., cost, carbon footprint, and worker injuries).

This process-based parametric modeling approach is capable of targeting individual processes, as well as addressing manufacturing systems composed of multiple processes. The approach enhances flexibility compared to traditional manufacturing assessment methods. Particularly, the approach is amenable to variations in system scope and boundaries (e.g., life cycle extent, processes types, and process input and output flows). Models can be created for a specific case or can utilize commercially available software databases to eliminate time-intensive and error-prone data collection. As another benefit, this approach enables the modeling and evaluation of economic, environmental, and social impacts simultaneously. When employed to evaluate multiple metrics under different scenarios, parametric modeling offers the opportunity for better understanding of system performance as discussed in Sec. 5. Herein, the approach is demonstrated for evaluating the environmental and social performance (e.g., energy use, carbon footprint, nonfatal occupational injuries and illnesses, and days away from work) of a metal-based product (bicycle pedal assembly).

## Methodology Overview

It is clear that sustainable manufacturing requires the integrated analysis of manufacturing processes and supply chain operations on the basis of associated economic, environmental, and social impacts at the product design stage. This section describes the methodology developed as a part of this research, as well as the assessment metrics and assessment framework.

The five primary steps of the methodology can be summarized as follows:

1. (1)

Identify potential material suppliers, their locations, and the mode of delivery to the manufacturer. This step involves identifying possible upstream supply chain networks and collecting related information, such as travel distance, transported material mass, and related environmental and social impact factors.

2. (2)

Identify potential component manufacturers, their locations, and the mode of delivery to the business or final customer. This step involves identifying possible midstream supply chain networks and collecting related information, as noted in step (1).

3. (3)

Characterize the environmental and social impacts of unit processes occurring at each node within the supply chain networks. For example, at each manufacturing facility identified in steps (1)–(2), unit process analysis would be conducted for a given product design to identify energy use, nonfatal occupational injuries, and days away from work. Equations related to calculation of social indicators and quantitative process models are described in Secs. 2.1 and 3, respectively.

4. (4)

Characterize the environmental and social impacts of the links between the nodes within the supply chain networks. For example, for each transportation path identified in steps (1)–(2), impacts on social metrics along with GHG emissions associated with fuel use to transport a given mass of material for a given distance would be calculated. Transportation models are described in Sec. 4.

5. (5)

Summarize and compare the impacts for the various manufacturing activities and transportation operations to select the most appropriate product, process, and supply chain to accomplish the desired objective.

### Assessment Metrics.

As mentioned above, the sustainability assessment metrics selected herein consider the environmental and social aspects (energy use, carbon footprint, nonfatal occupational injuries and illnesses, and days away from work), since economic aspects are of traditional focus within engineering decision making. Further, the methodology introduced above is amenable to inclusion of a variety of metrics. Based on the importance of considering social impacts over the product life cycle, guidelines such as the Global Reporting Initiative (GRI) framework and the ISO 26000 standard have been developed. In addition, social impact databases have emerged, such as the Social Hotspots Database (SHD) and the Fair Factories Clearinghouse (FFC) to aid decision makers. Supporting methodologies and case studies also have been developed [76,77]. Three social issues of common concern in the supply chain are labor conditions, health, and safety [14]. Consequently, related performance metrics are selected to quantify social sustainability in this research, i.e., nonfatal occupational injuries and illnesses (NOI) and days away from work (DAW). The number of nonfatal occupational injuries and illnesses indicates the level of safety within the work environment. The measure is based on data commonly collected by companies and available for the U.S. from the Bureau of Labor Statistics (BLS) [78]. To calculate the number of nonfatal occupational injuries and illnesses for a given production volume processed using a given unit operation (Eq. (1)), the nonfatal occupational injury rate for the relevant industry segment provided by the BLS is multiplied by process time and production volume. This quantity is divided by 200,000 h, which is the assumed annual working hours for 100 equivalent full-time workers
$NOImfg=RNOImfg⋅(Tmfg⋅PV200,000)$
(1)
Some injuries and illnesses cause lost working time, captured by the BLS as the rate of injuries and illnesses resulting in days away from work. The amount of lost time is dependent on the severity of the injury and is reported for different industries by the BLS. To calculate the number of days away from work (DAW) for a given manufacturing process, the percentage of injuries and illnesses that result in days away from work is first determined from BLS data. This value is multiplied by NOImfg to provide the total number of cases resulting in days away from work. Finally, the product of this number and the median days away from work reported for the relevant industry segment provides the total number of days away from work for the process (Eq. (2))
$DAWmfg=(RDAWmfgRNOImfg)⋅NOImfg⋅MDAWmfg$
(2)
Equations (3) and (4) present NOI and DAW for transportation. The calculation process is similar to that for manufacturing, only differing in the calculation of NOI. Rather than process time, the result from dividing transportation time by the number of parts that can be transported by the vehicle is considered
$NOItran=RNOItran⋅((TtranNtran)⋅PV200,000)$
(3)
$DAWtran=(RDAWtranRNOItran)NOItran⋅MDAWtran$
(4)

How to best measure social metrics is often uncertain. This uncertainty stems from the variable perceptions of social impacts and the mix between qualitative, semiquantitative, and quantitative measurements [11]. Although developing social metrics is a challenging task, it is essential to make design and manufacturing engineers aware of the social impacts of their work and decisions [16]. The selected social metrics in this research are commonly understood, are applicable to various companies, and are easily measurable. Other metrics such as employee turnover, employee empowerment, working conditions, work's impact on worker's long-term health, and chronic illnesses could be identified and applied for assessing social impacts if desired [16,21].

The environmental impact indicators considered herein are energy use and carbon footprint, which are both related to human activities. For example, the examined transportation and manufacturing processes are energy-driven and currently rely heavily on nonrenewable sources. Carbon footprint measures the impact of different energy sources based on the distribution of various GHG emissions. Fossil fuels are the basis of electricity generation and direct industrial and transportation energy consumption [79]. Thus, industrial and transportation energy use strongly influences sector-related carbon dioxide (CO2) and other GHG emissions. GHGs have the potential to absorb infrared radiation, trapping heat in the lowest layers of the atmosphere [80,81], which raises concerns about their influence on the global environment.

The concept of carbon footprint has been introduced to provide a common measure of overall GHG emissions, as well as providing a metric for tracking improvement [82]. Carbon footprint is provided as an equivalent mass of CO2 for GHGs that are directly and indirectly caused by an activity or accumulated over the life stages of a product [83]. The importance of using carbon footprint as an environmental measure of manufacturing activities has been reported widely in the literature [53,54,84,85] and is adopted by this research as a means of demonstrating the methodological concepts. A key focus in this work is in modeling of manufacturing process and supply chain impacts, as discussed in Secs. 2.2 and 2.3.

### Assessment Framework.

The approach proposed herein for environmental and social impact assessment accounts for various factors to support simultaneous design of manufacturing process flows and the supply chain network within a cradle-to-gate scope (Fig. 1). The figure shows this scope—highlighted in the darker gray box, i.e., raw material extraction, material processing, and manufacturing— in the context of the cradle-to-grave product life cycle, which consists of stages from material extraction through end-of-life. Product design is illustrated as an input that defines the required activities across all phases. The actual impacts of the different phases remain unknown until all activities are complete.

Fig. 1
Fig. 1
Close modal

Data and information must be collected for each process from cradle to gate to evaluate selected metrics in a simultaneous, process-based manner. Thus, the energy use, carbon footprint, nonfatal occupational injuries and illnesses, and days away from work can be calculated for various supply chain configurations to determine the preferable route for product manufacture. Figure 2 shows a network of potential supply chain configurations for the production of n components of a given product. Each supply chain configuration is subject to a combination of different transportation modes, supplier locations, and routes (distances). If a supply chain network is selected as indicated in the figure, for example, transportation-related impacts can be calculated by summing the resultant impacts for each link (connecting line), while manufacturing-related impacts would be calculated by summing the impacts for each node.

Fig. 2
Fig. 2
Close modal

To evaluate impacts of component production, the energy, materials, and cycle time can be determined for each process (Fig. 3). Each input and output represents a vector of flows, which must be normalized to a common reference flow. While this procedure is well established for LCA, it is data, time, and resource intensive. Thus, to facilitate the approach, process-based models that are indicative of each process can be devised that elucidate the mathematical relationships of process inputs and outputs [24]. Similarly, relationships can be established for supply chain entities, describing transportation modes, distances, and associated performance metrics on a mass or volume basis. Product types can be mass or volume limited for transport using common modes.

Fig. 3
Fig. 3
Close modal

In addition to identifying the best supply chain configuration, the analysis would contribute to targeting the most impactful entities and facilitate reconfiguration of the supply chain. If existing supply chains are evaluated and found to perform poorly from a sustainability perspective, alternative manufacturing process and supply chain solutions could be developed that lead to process innovations or identification of new supply chain partners.

## Manufacturing Energy Models

Product materials undergo mechanical work and chemical reactions using unit manufacturing processes in an effort to amend one or several characteristics of its original stage [86]. Materials, chemicals, electricity, and fuels are primary inputs at the process level. In this section, development of energy consumption models for various processes is provided. Forging, laser cutting, bending, and machining are selected manufacturing processes for investigation for two main reasons: (1) they are deemed to be the largest energy users in the production of the components under study herein, and (2) energy data and/or energy prediction models exist in the literature to support metric quantification. All model parameters are defined in the Nomenclature section, while process efficiency factors and key model parameter values are provided in Tables 1 and 2, respectively.

Table 1

Process efficiency factors

Manufacturing processEfficiency factorSource
Bulk metal deformation (drawing, extrusion, and forging)0.60Ref. [87]
Casting0.55Ref. [88]
Laser cutting0.15–0.25 (avg. 0.20)Ref. [89]
Machining process (drilling, milling, and chamfering)0.65–0.70 (avg. 0.67)Ref. [90]
Sheet metal working (bending, punching, and stamping)0.67Ref. [91]
Turning/boring0.75Ref. [92]
Tapping0.77Ref. [93]
Manufacturing processEfficiency factorSource
Bulk metal deformation (drawing, extrusion, and forging)0.60Ref. [87]
Casting0.55Ref. [88]
Laser cutting0.15–0.25 (avg. 0.20)Ref. [89]
Machining process (drilling, milling, and chamfering)0.65–0.70 (avg. 0.67)Ref. [90]
Sheet metal working (bending, punching, and stamping)0.67Ref. [91]
Turning/boring0.75Ref. [92]
Tapping0.77Ref. [93]
Table 2

Key model parameter values

Unit manufacturing processParameterValueUnitSource
BendingStrength coefficienta965MPaRef. [94]
Strain-hardening exponenta0.14
Bulk metal deformation (drawing and forging)Coefficient of frictiona0.25Ref. [94]
Strength coefficienta965MPa
Strain-hardening exponenta0.14
CastingEnergy use per unit (EAF)10,990kJ/kgRef. [95]
Laser cuttingHeat source density1000kW/cm2Ref. [96]
Material constantb5000W/cm2/s
Machining process (boring, drilling, milling, chamfering, and tapping)Average unit power requirement0.064kW/cm3/minRef. [94]
Sheet metal working (punching and stamping)Ultimate tensile strength579MPaRef. [97]
Stock metal productionUnit energy use (EAF)11,500kJ/kgRef. [95]
Unit manufacturing processParameterValueUnitSource
BendingStrength coefficienta965MPaRef. [94]
Strain-hardening exponenta0.14
Bulk metal deformation (drawing and forging)Coefficient of frictiona0.25Ref. [94]
Strength coefficienta965MPa
Strain-hardening exponenta0.14
CastingEnergy use per unit (EAF)10,990kJ/kgRef. [95]
Laser cuttingHeat source density1000kW/cm2Ref. [96]
Material constantb5000W/cm2/s
Machining process (boring, drilling, milling, chamfering, and tapping)Average unit power requirement0.064kW/cm3/minRef. [94]
Sheet metal working (punching and stamping)Ultimate tensile strength579MPaRef. [97]
Stock metal productionUnit energy use (EAF)11,500kJ/kgRef. [95]
a

For SAE1045, hot rolled steel.

b

Function of thermal conductivity and thermal diffusivity of the material (steel).

### Forging.

Forging is a deformation process that applies compressive forces on a workpiece using open or closed dies to form geometric features. The major application of forging is the manufacturing of discrete parts, such as bolts, nuts, rivets, and steel balls [94]. The compressive force used in an open-die forging process is shown in the following equation [94]:
$FF=Yfπr2(1+2μr3h)$
(5)
Flow stress, $Yf$, can be obtained for various materials based on given curves of true strain, $εf$, versus true stress, $σT$ [94]. The actual power can then be calculated for a given efficiency factor, $ηf$, between 0 and 1 and a die speed, $vf$, capable of achieving the required force as given in the following equation:
$Pactual=FFvf(1+ηf)$
(6)
Thus, the energy consumption can be computed according to Eq. (7) in which the die speed and the final workpiece height, $hf$, are available
$ECF=Pactualhfvf$
(7)

### Laser Cutting.

One of the most commonly used processes in plate cutting is a laser cutting. It provides a narrow kerf width and high finish precision due to computerized control. Laser cutting is able to process a variety of thickness and shapes. Another advantage is the ability to combine laser cutting with other processes like punching and shearing. The relationship for energy consumption in laser cutting is presented as follows:
$ECL=P⋅X⋅ηlvl$
(8)

### Bending.

Bending is a deformation process in which the workpiece is manipulated on the basis of a neutral axis at different bending angles according to part design specifications and limited by the bend allowance. Sheet metals most commonly undergo bending. Forces are compressive on the inside of the part's neutral plane, while tensile forces act outside of the part's neutral plane. The bending force can be described as follows [98]:
$FB=Y⋅t2⋅ cos(a2).(cos(a2)+μ⋅ sin(a2))Wd−(2⋅ sin(a2).(R sin a+t))+μ⋅t⋅ cos(a2)$
(9)
The flow stress, Y, is a function of the punch radius, R, the workpiece thickness, t, and bend angle, a, as presented in the following equation [98]:
$Y=2.S3(e+1).(log 1+t⋅ sin aR)e$
(10)

For simplicity, the energy consumption can be calculated by means of Simpson's 3/8 rule for numerical integration once the bending force and punch displacement are calculated for various bending angles. This can be performed using a spreadsheet program to obtain the area under the curve for corresponding bending forces at various punch displacements [96]. The bending process efficiency factor is provided in Table 1.

### Machining.

Machining consists of several material removal processes (e.g., milling, turning, and drilling). Milling involves material removal from a stationary workpiece using a rotating tool. In turning, a rotating workpiece is cut with a single cutting edge moving parallel to the workpiece rotational axis. Turning processes can take on various forms (e.g., grooving, boring, and threading). The drilling process is used to create through and blind holes, typically using a helical drill bit. The following equations give the energy consumption for turning and drilling processes [94]:
$ECT=π(Di2−Df24)LT×U×ηt$
(11)
$ECD=πD024d×U×ηd$
(12)

## Supply Chain Energy Models

The environmental performance of the supply chain transportation is affected by several factors such as vehicle technology (e.g., conventional vehicles with internal combustion engines, electric vehicles, and hybrid electric vehicles), transportation network efficiency, and types of fuel. For instance, diesel produces greater GHG emissions than biodiesel [99].

It can be assumed that starting location of each supply chain, the first node, represents the location at which all material upstream processing occurs. By the same token, it can be assumed that the n − 1 nodes represent all remaining manufacturing process locations for a part. Assembly of components will occur at location n. The GHG emissions due to transportation from location j − 1 to location j, et(j), can be expressed as the product of the mass to be transported, m, the distance of the travel link, dj, and the emission factor of the transportation mode utilized, βm, as expressed in Eq. (13). Travel links are based on a starting point, j − 1, and ending point, j, where j = 1, 2,…, l
$et(j)=m⋅dj⋅βm$
(13)
Similarly, GHG emissions due to manufacturing process i at location j, em(i,j), is the product of process energy consumption, E(i,j), and the energy to carbon footprint conversion factor, αe, for location j, as shown in the following equation:
$em(i,j)=E(i,j)⋅αe$
(14)
For example, the energy to carbon footprint conversion factor can be obtained from the eGrid database developed by the U.S. EPA [100] or from the U.S. EIA dataset, which also includes data from foreign countries [101]. The following equation provides total emissions from manufacturing and transportation activities:
$etotal=∑jm−1et(j)+∑i=1nem(i,j)$
(15)

Table 3 provides the average emission factors, βm, for various means of transportation.

Table 3

Recommended average emission factors [102]

Transportation modeEmission factor (g CO2 eq./tonne km)
Rail transport22
Barge transport31
Short sea16
Deep-sea container8
Deep-sea tanker5
Air freight602
Transportation modeEmission factor (g CO2 eq./tonne km)
Rail transport22
Barge transport31
Short sea16
Deep-sea container8
Deep-sea tanker5
Air freight602

## Demonstration of the Framework

This section presents an application of the method developed in Secs. 3 and 4 using a bicycle pedal. Manufacturing process flows required for the manufacture of each pedal component were identified and analyzed using the sustainability metrics identified above: energy use, carbon footprint, nonfatal occupational injuries and illnesses, and days away from work. Similarly, supply chain network design alternatives were assessed. This work builds upon a prior study by the authors [96].

### Representative Product.

The bicycle pedal considered (Fig. 4) is assembled from ten primary components: nine parts are assumed to be made of carbon steel and one is made of an aluminum alloy. Figure 5 shows the disassembled product components with the corresponding identification numbers and names. The identification numbers are used here to refer to the different parts. Supply chain scenarios for each of the ten components were constructed and analyzed using the procedure described in Sec. 4. Next, the methodology introduced in Sec. 2 is illustrated for the body plate (part 3). Results are then summarized for the pedal assembly.

Fig. 4
Fig. 4
Close modal
Fig. 5
Fig. 5
Close modal

### Manufacturing and Supply Chain Analysis.

Supplier locations were selected based on common production locations for the required raw materials or process type. For instance, steel suppliers could be from China, India, the U.S., or Luxembourg. Similarly, the selection of the links and routes are based on geographical location related to the final destination. Therefore, the nodes (i.e., suppliers–suppliers, warehouses, and retailers) were chosen and defined based on distance calculators developed by Google [103] and NASA [104]. The final destination (assembly location) was chosen to be Irvine, CA. Table 4 shows the alternative supply chains scenarios for steel and aluminum components. Each supply chain in the table is assumed to cover the cradle-to-gate life cycle scope. Locations for material processing and manufacturing processes are shown along with the transportation modes utilized and relevant distances. Two supply chain scenarios for each of the ten components are shown and labeled “a” or “b.”

Table 4

Supply chain alternatives

MaterialSC (codea)FromToModeDistance (km)
Steel alloys1 (2a, 3a, 4a, 6a, 7a, 10a)Luxembourg, LuxembourgCardiff, Wales, UKRail691
Cardiff, Wales, UKBridgeport, CTDeep-sea container5297
2 (1a, 3b, 5a, 7b, 9a)New Delhi, IndiaLondon, UKDeep-sea container6724
London, UKNew York, NYBarge5587
3 (1b, 4b, 5b)Pittsburg, PAIrvine, CAIntermodal road/rail3417
4 (2b, 6b, 9b, 10b)Beijing, ChinaShanghai, ChinaIntermodal road/rail1066
Shanghai, ChinaHonolulu, HIDeep-sea container7964
Aluminum alloys1 (8a)New York, NYPittsburg, PARoad508
Boston, MASalt Lake City, UTAir freight3376
Salt Lake City, UTRiverside, CARail902
MaterialSC (codea)FromToModeDistance (km)
Steel alloys1 (2a, 3a, 4a, 6a, 7a, 10a)Luxembourg, LuxembourgCardiff, Wales, UKRail691
Cardiff, Wales, UKBridgeport, CTDeep-sea container5297
2 (1a, 3b, 5a, 7b, 9a)New Delhi, IndiaLondon, UKDeep-sea container6724
London, UKNew York, NYBarge5587
3 (1b, 4b, 5b)Pittsburg, PAIrvine, CAIntermodal road/rail3417
4 (2b, 6b, 9b, 10b)Beijing, ChinaShanghai, ChinaIntermodal road/rail1066
Shanghai, ChinaHonolulu, HIDeep-sea container7964
Aluminum alloys1 (8a)New York, NYPittsburg, PARoad508
Boston, MASalt Lake City, UTAir freight3376
Salt Lake City, UTRiverside, CARail902
a

Alternate scenarios coded for each part number.

Part 3, the body plate (Fig. 6) is assumed to be constructed from 11 gauge (3 mm thick) stainless steel sheet and is 59.9 g in mass. The blank from which the part is cut is assumed to have a mass of 135 g.

Fig. 6
Fig. 6
Close modal

The pedal body plate can be manufactured in a number of ways, as dictated by the design and supply chain capabilities. Typically, it would undergo steelmaking, continuous casting, rolling, punching, bending, and machining processes. Two alternative scenarios are presented in Fig. 7. Scenario SC3a is capable of producing the part in the following sequence: stock sheet production, laser cutting, bending, and chamfering. Luxembourg is assumed as the steel sheet production location, which would utilize an electric arc furnace (EAF) mini-mill, and Columbus, OH is assumed as the body plate manufacturing facility.

Fig. 7
Fig. 7
Close modal

Scenario SC3b imposes a different manufacturing process flow (e.g., stock sheet production, stamping, bending, and chamfering) due to supply chain capabilities, as shown in Fig. 7 (down). New Delhi, India is the production location for the steel plate, assumed to use a basic oxygen furnace (BOF) integrated mill. Body plate component production will occur in Austin, TX. Other locations indicate material/part transfer points in the supply chain and the final destination is Irvine, CA, where assembly of the product occurs.

The basis of the analysis is one pedal body plate. The carbon footprint due to transportation can be calculated using Eq. (1), where the volume to be reported from Luxembourg, Luxembourg to Columbus, OH is the blank mass from which the actual product is made (135 g), for which the mass is determined by multiplying the part bounding box volume by the material density. Table 5 provides additional information about the supply chain locations, mass of the part at the upstream locations, energy consumption during manufacturing processes, relevant GHG emission factors, and carbon footprint at each step.

Table 5

Supply chain scenarios for manufacturing of the pedal body plate (part 3) [35,105,106]

SCManufacturing locationMultiplierEmission factorCarbon footprint (kg CO2 eq.)
3aLuxembourg, Luxembourg0.132 (kg)2.9 × 10−1 (kg CO2 eq./kg)0.03828
Columbus, OH86.35 (kJ)19.7 × 10−5 (kg CO2 eq./kJ)0.01701
3bNew Delhi, India0.132 (kg)4.7 × 10−1 (kg CO2 eq./kg)0.06204
Austin, TX46.11 (kJ)15.8 × 10−5 (kg CO2 eq./kJ)0.00728
SCManufacturing locationMultiplierEmission factorCarbon footprint (kg CO2 eq.)
3aLuxembourg, Luxembourg0.132 (kg)2.9 × 10−1 (kg CO2 eq./kg)0.03828
Columbus, OH86.35 (kJ)19.7 × 10−5 (kg CO2 eq./kJ)0.01701
3bNew Delhi, India0.132 (kg)4.7 × 10−1 (kg CO2 eq./kg)0.06204
Austin, TX46.11 (kJ)15.8 × 10−5 (kg CO2 eq./kJ)0.00728

The transported mass from Columbus, OH to Irvine, CA is equal to the final mass of the product (59.9 g), since it has undergone all material removal processes. The GHG emission factors are obtained from Table 1 based on transportation mode utilized. The carbon footprint due to manufacturing processes was calculated using Eq. (2) after computing energy use for each unit manufacturing process adopted in the product manufacturing sequence. The energy use in the upstream process captures the total energy from material extraction, transportation, and associated services in the making of the material. All unit processes require knowledge about the part information (e.g., dimensional characteristics and material type), as well as machine details to calculate energy use as reported by Alsaffar [96]. The energy use is then converted to carbon footprint using common conversion factors [35]. The results are summarized in Fig. 8 for pedal body plate production supply chain alternatives SC3a and SC3b, respectively.

Fig. 8
Fig. 8
Close modal

It is clear that the manufacturing processes dominate in SC3a and transportation activities are the greater contributor to cradle-to-gate carbon footprint in SC3b with steel production having the highest manufacturing-related carbon footprint for both scenarios.

#### Environmental Assessment of Supply Chain Networks.

Analyses pertaining to supply chain network design are performed in this section. Figures 9 and 10 show the cumulative distances, transportation modes, and carbon footprints for alternatives SC3a and SC3b. In SC3a, the total distance is 9991 km for which around 53% is via sea using a deep-sea container. It is apparent that the resultant carbon footprint from Cardiff, Wales, UK to Bridgeport, CT is lower than other transportation modes relevant to the traveled distance. In fact, the total GHG emissions generated using the deep-sea container accounts for one-fifth of the total carbon footprint. In comparison, SC3b requires a total distance of 16,671 km traveled, and its carbon footprint is disproportionately higher due to the use of air freight.

Fig. 9
Fig. 9
Close modal
Fig. 10
Fig. 10
Close modal

The foregoing shows that road transport dominates the transportation carbon footprint in SC3a, accounting for 71% of the impact, while air freight accounts for 91% transportation carbon footprint of SC3b (Table 6). Each comprises only 40% of the total distance traveled in their respective supply chains. Thus, these transportation routes could be identified as impactful to the supply chain carbon footprint outcome. Prior knowledge of the nature of the routes would be necessary in identifying alternative transportation modes, such as rail or intermodal rail/road transportation, which can alleviate the production of GHG emissions.

Table 6

Distance and carbon footprint (CF) contributions by transportation mode for supply chain alternatives SC3a and SC3b

SC3aSC3b
Transport modeDist. (%)CF (%)Dist. (%)CF (%)
Rail6.927.65
Deep-sea container53.0221.32
Air freight40.3391.45
Barge33.513.91
SC3aSC3b
Transport modeDist. (%)CF (%)Dist. (%)CF (%)
Rail6.927.65
Deep-sea container53.0221.32
Air freight40.3391.45
Barge33.513.91

SC3b involves about 67% more travel distance compared to SC3a and has about 20 times greater carbon footprint than SC3a due to dependence on air transport by air freight. Hence, transportation planning could be exploited to replace impactful transportation modes and, subsequently, enhance supply chain environmental performance. The observations noted in this section will become particularly useful when considered in conjunction with the manufacturing process environmental performance as will be seen in Sec. 5.2.2.

#### Environmental Assessment of Manufacturing Processes.

Figure 11 shows the resultant carbon footprint from manufacturing activities associated with SC3a and SC3b. Black bars show the carbon footprint from upstream processes, while white bars show the total carbon footprint from cutting, forming, and machining processes. As previously discussed, the impacts of upstream processes are affected by the type of steelmaking used and the source profile of electrical energy. It can be noted that the upstream processing in SC3a exhibits a 40% lower carbon footprint than those that in SC3b due to the use of the electric arc furnace (EAF), rather than the basic oxygen furnace (BOF). While the upstream carbon footprint is lower than that in SC3b, forming and machining process in SC3a generate a carbon footprint 2.5 times greater than manufacturing processes in SC3b. Thus, further investigation is imperative in identifying and eliminating impactful processes.

Fig. 11
Fig. 11
Close modal

Figure 12 shows the energy consumption and related carbon footprint for each unit manufacturing process within the forming and machining categories for alternatives SC3a and SC3b, respectively. From the figures, there are two main observations regarding the amount of GHG emissions from each process flow. First, while chamfering and bending are common processes between the two flows and they consume the same energy in each scenario, their carbon footprints differ due to the electrical energy profile of each location. The second observation is regarding the alternative cutting processes used in each sequence (i.e., laser cutting and stamping). Laser cutting consumes around 80 kJ, in contrast to stamping which consumes around 40 kJ. Hence, it can be identified as an impactful process prior to carbon footprint conversion.

Fig. 12
Fig. 12
Close modal

Figure 13 illustrates the process sequence and associated energy and carbon footprint for each supply chain. It can be seen, while the required energy does not vary, the carbon footprint of forming and machining processes in SC3a is greater than SC3b. The GHG emission factor in Ohio, which has coal energy generation, is 0.197 × 10−3 g CO2 eq./kJ. It is fairly similar to 0.158 × 10−3 g CO2 eq./kJ in Texas from natural gas energy generation. Since stamping consumes less energy than laser cutting, however, it could be a potential alternative to laser cutting despite the slightly higher impact associated with the electrical energy profile of Ohio.

Fig. 13
Fig. 13
Close modal

The carbon footprint results for both alternatives are displayed in Fig. 14. The impact of SC3b is greater primarily due to the use of air freight and higher upstream processing in India with use of a BOF.

Fig. 14
Fig. 14
Close modal

It can be seen that the transportation impact in terms of carbon footprint greatly surpasses that for manufacturing. Summary of the environmental results for SC3a and SC3b is reported in Table 7.

Table 7

SC3a and SC3b environmental results summary

SubjectEntity descriptionSC3aSC3b
TransportationTotal travel distance (km)999116,671
Carbon footprint from sea travel (kg CO2 eq.)5.59 × 10−32.29 × 10−2
Carbon footprint from rail travel (kg CO2 eq.)2.01 × 10−30
Carbon footprint from road travel (kg CO2 eq.)1.86 × 10−22.71 × 10−2
Carbon footprint from air travel (kg CO2 eq.)05.34 × 10−1
Total carbon footprint from transportation (kg CO2 eq.)2.62 × 10−25.84 × 10−1
ManufacturingCarbon footprint from upstream processes (kg CO2 eq.)3.83 × 10−26.2 × 10−2
Carbon footprint from manufacturing processes (kg CO2 eq.)1.70 × 10−27.31 × 10−3
Total carbon footprint from manufacturing (kg CO2 eq.)5.53 × 10−26.93 × 10−2
FactorsUpstream location emission factor (kg CO2 eq./kg)0.290.47
Manufacturing location emission factor (kg CO2 eq./kJ)1.965 × 10−41.584 × 10−4
Supply chainTotal carbon footprint (kg CO2 eq.)8.15 × 10−26.54 × 10−1
SubjectEntity descriptionSC3aSC3b
TransportationTotal travel distance (km)999116,671
Carbon footprint from sea travel (kg CO2 eq.)5.59 × 10−32.29 × 10−2
Carbon footprint from rail travel (kg CO2 eq.)2.01 × 10−30
Carbon footprint from road travel (kg CO2 eq.)1.86 × 10−22.71 × 10−2
Carbon footprint from air travel (kg CO2 eq.)05.34 × 10−1
Total carbon footprint from transportation (kg CO2 eq.)2.62 × 10−25.84 × 10−1
ManufacturingCarbon footprint from upstream processes (kg CO2 eq.)3.83 × 10−26.2 × 10−2
Carbon footprint from manufacturing processes (kg CO2 eq.)1.70 × 10−27.31 × 10−3
Total carbon footprint from manufacturing (kg CO2 eq.)5.53 × 10−26.93 × 10−2
FactorsUpstream location emission factor (kg CO2 eq./kg)0.290.47
Manufacturing location emission factor (kg CO2 eq./kJ)1.965 × 10−41.584 × 10−4
Supply chainTotal carbon footprint (kg CO2 eq.)8.15 × 10−26.54 × 10−1

Figure 15 shows the carbon footprint of the transportation and manufacturing activities for all pedal components. It is clear that the manufacturing activities dominate the production of GHG emissions for most components. SC7b uses air freight transport and dominates the generation of GHG emissions compared to SC7a, which utilizes deep-sea container shipping as the main transportation mode. The remarkable difference between SC8a and SC8b in carbon footprint from transportation activities is because SC8a utilizes only truck transport while SC8b uses air freight transport.

Fig. 15
Fig. 15
Close modal

Figure 16 presents the carbon footprint produced in the best and worst scenarios related to each component for manufacturing and transportation activities, respectively. These figures are normalized based on the weight of each component.

Fig. 16
Fig. 16
Close modal

Next, the pedal components and assemblies with the largest and smallest carbon footprints are considered for each individual scenario. Figure 17 shows the best and worst alternatives for each component and is normalized by the weight of each component (the impact is reported on a kg CO2 eq. per kg of product basis). In this manner, the method can aid in identifying the most impactful activities in the supply chain to reduce carbon footprint.

Fig. 17
Fig. 17
Close modal

Figure 18 shows the environmental performance for the pedal assembly in terms of carbon footprint, demonstrating the best and worst options. It can be seen that the worst assembly has a carbon footprint around 3.5 times greater than the best assembly.

Fig. 18
Fig. 18
Close modal

The analyses conducted show that the simultaneous consideration of manufacturing and transportation activities can impact the product environmental performance at the design development stage. Section 5.2.3 explores the social performance assessment.

#### Social Performance Assessment.

From a manufacturing perspective, process time is an important factor that correlates with environmental and social impacts. Considering other social metrics, such as working hours, labor equity, and workload, would further emphasize the important role of process time in sustainability analysis. The focus of manufacturing process evaluation herein is on nonfatal occupational injuries and illnesses (NOI) and days away from work (DAW). Other social metrics, such as child labor, raw material extraction from disadvantaged regions, and level of community involvement, are not directly related to process time, but can be key manufacturing sustainability indicators.

Figures 19 and 20 display the social impact analysis results for transportation and manufacturing processes for one pedal under supply chain alternatives SC3a and SC3b, respectively. Shorter manufacturing times in SC3a result in a lower NOI. Since DAW is calculated based on NOI and associated severity rates, a lower DAW value is also found for SC3a. The severity of injuries and illnesses is assumed to be similar across manufacturing process types based on U.S. BLS data; however, higher fidelity data (e.g., company-specific injury data) could elucidate variations in severity and lead to higher variations in DAW across the supply chain alternatives.

Fig. 19
Fig. 19
Close modal
Fig. 20
Fig. 20
Close modal

Section 5.3 summarizes the results of the environmental and social impact analysis.

### Summary of Results.

Tables 8 and 9 are presented to summarize the foregoing analysis. The tables demonstrate that scenarios defined as SCa consume less energy for manufacturing and transportation for most parts compared to SCb scenarios. Consequently, they also have a lower carbon footprint. The differences between these scenarios when considering manufacturing energy consumption and carbon footprint are 74% and 23%, respectively. Transportation activities under SCb scenarios have an energy consumption and carbon footprint of about seven times more than SCa scenarios. Evaluating transportation and manufacturing carbon footprint on the basis of a complete pedal assembly reveals that the manufacturing carbon footprint surpassed that of transportation. This indicates that consideration of unit manufacturing processes in the supply chain is critical to design and manufacturing decision making.

Table 8

SC3a and SC3b environmental and social results summary from manufacturing processes (light shading indicates highest impact and dark shading indicates lowest impact)

Energy consumption (kJ)Carbon footprint (kg CO2 eq.)Nonfatal occupational injuries and illnessesDays away from work
Part no.SCaSCbSCaSCbSCaSCbSCaSCb
12.42 × 10−12.42 × 10−17.38 × 10−44.79 × 10−47.46 × 10−107.46 × 10−101.57 × 10−101.57 × 10−10
22.38 × 1012.38 × 1016.21 × 10−34.53 × 10−33.17 × 10−53.17 × 10−51.00 × 10−41.00 × 10−4
38.63 × 1014.61 × 1015.53 × 10−26.93 × 10−25.34 × 10−46.23 × 10−71.69 × 10−31.97 × 10−6
41.18 × 1011.18 × 1013.08 × 10−33.08 × 10−33.17 × 10−53.17 × 10−51.00 × 10−41.00 × 10−4
53.754.948.40 × 10−35.79 × 10−32.04 × 10−42.03 × 10−46.44 × 10−46.42 × 10−4
67.749.671.09 × 10−21.60 × 10−22.68 × 10−42.67 × 10−48.46 × 10−48.44 × 10−4
71.43 × 1011.09 × 1039.11 × 10−31.83 × 10−12.64 × 10−52.64 × 10−58.33 × 10−58.33 × 10−5
81.12 × 1031.15 × 1033.03 × 10−11.81 × 10−11.85 × 10−41.85 × 10−45.84 × 10−45.84 × 10−4
91.47 × 1021.952.43 × 10−21.29 × 10−33.09 × 10−93.09 × 10−99.76 × 10−99.76 × 10−9
102.791.41 × 1021.05 × 10−11.82 × 10−13.60 × 10−53.88 × 10−41.14 × 10−41.23 × 10−3
Energy consumption (kJ)Carbon footprint (kg CO2 eq.)Nonfatal occupational injuries and illnessesDays away from work
Part no.SCaSCbSCaSCbSCaSCbSCaSCb
12.42 × 10−12.42 × 10−17.38 × 10−44.79 × 10−47.46 × 10−107.46 × 10−101.57 × 10−101.57 × 10−10
22.38 × 1012.38 × 1016.21 × 10−34.53 × 10−33.17 × 10−53.17 × 10−51.00 × 10−41.00 × 10−4
38.63 × 1014.61 × 1015.53 × 10−26.93 × 10−25.34 × 10−46.23 × 10−71.69 × 10−31.97 × 10−6
41.18 × 1011.18 × 1013.08 × 10−33.08 × 10−33.17 × 10−53.17 × 10−51.00 × 10−41.00 × 10−4
53.754.948.40 × 10−35.79 × 10−32.04 × 10−42.03 × 10−46.44 × 10−46.42 × 10−4
67.749.671.09 × 10−21.60 × 10−22.68 × 10−42.67 × 10−48.46 × 10−48.44 × 10−4
71.43 × 1011.09 × 1039.11 × 10−31.83 × 10−12.64 × 10−52.64 × 10−58.33 × 10−58.33 × 10−5
81.12 × 1031.15 × 1033.03 × 10−11.81 × 10−11.85 × 10−41.85 × 10−45.84 × 10−45.84 × 10−4
91.47 × 1021.952.43 × 10−21.29 × 10−33.09 × 10−93.09 × 10−99.76 × 10−99.76 × 10−9
102.791.41 × 1021.05 × 10−11.82 × 10−13.60 × 10−53.88 × 10−41.14 × 10−41.23 × 10−3
Table 9

SC3a and SC3b environmental and social results summary from transportation (light shading indicates highest impact and dark shading indicates lowest impact)

Energy consumption (kJ)Carbon footprint (kg CO2 eq.)Nonfatal occupational injuries and illnessesDays away from work
Part no.SCaSCbSCaSCbSCaSCbSCaSCb
19.40 × 10−53.01 × 10−66.61 × 10−31.32 × 10−43.13 × 10−71.94 × 10−94.47 × 10−62.81 × 10−8
21.22 × 10−51.18 × 10−58.73 × 10−48.60 × 10−44.45 × 10−72.26 × 10−76.27 × 10−63.08 × 10−6
33.68 × 10−48.30 × 10−32.62 × 10−25.84 × 10−11.43 × 10−52.60 × 10−52.02 × 10−43.72 × 10−4
47.35 × 10−63.02 × 10−65.24 × 10−41.07 × 10−42.86 × 10−73.12 × 10−94.04 × 10−64.51 × 10−8
51.05 × 10−33.29 × 10−57.38 × 10−21.48 × 10−33.45 × 10−62.13 × 10−84.92 × 10−53.08 × 10−7
69.09 × 10−57.25 × 10−56.48 × 10−35.27 × 10−33.55 × 10−61.39 × 10−65.01 × 10−51.89 × 10−5
74.93 × 10−51.36 × 10−33.53 × 10−39.58 × 10−21.78 × 10−63.94 × 10−62.51 × 10−55.62 × 10−5
83.41 × 10−49.29 × 10−32.40 × 10−26.52 × 10−11.75 × 10−54.46 × 10−62.48 × 10−47.21 × 10−5
91.51 × 10−45.38 × 10−61.06 × 10−23.91 × 10−44.81 × 10−71.03 × 10−76.87 × 10−61.40 × 10−6
107.46 × 10−48.10 × 10−45.34 × 10−25.88 × 10−22.58 × 10−51.54 × 10−53.63 × 10−42.10 × 10−4
Energy consumption (kJ)Carbon footprint (kg CO2 eq.)Nonfatal occupational injuries and illnessesDays away from work
Part no.SCaSCbSCaSCbSCaSCbSCaSCb
19.40 × 10−53.01 × 10−66.61 × 10−31.32 × 10−43.13 × 10−71.94 × 10−94.47 × 10−62.81 × 10−8
21.22 × 10−51.18 × 10−58.73 × 10−48.60 × 10−44.45 × 10−72.26 × 10−76.27 × 10−63.08 × 10−6
33.68 × 10−48.30 × 10−32.62 × 10−25.84 × 10−11.43 × 10−52.60 × 10−52.02 × 10−43.72 × 10−4
47.35 × 10−63.02 × 10−65.24 × 10−41.07 × 10−42.86 × 10−73.12 × 10−94.04 × 10−64.51 × 10−8
51.05 × 10−33.29 × 10−57.38 × 10−21.48 × 10−33.45 × 10−62.13 × 10−84.92 × 10−53.08 × 10−7
69.09 × 10−57.25 × 10−56.48 × 10−35.27 × 10−33.55 × 10−61.39 × 10−65.01 × 10−51.89 × 10−5
74.93 × 10−51.36 × 10−33.53 × 10−39.58 × 10−21.78 × 10−63.94 × 10−62.51 × 10−55.62 × 10−5
83.41 × 10−49.29 × 10−32.40 × 10−26.52 × 10−11.75 × 10−54.46 × 10−62.48 × 10−47.21 × 10−5
91.51 × 10−45.38 × 10−61.06 × 10−23.91 × 10−44.81 × 10−71.03 × 10−76.87 × 10−61.40 × 10−6
107.46 × 10−48.10 × 10−45.34 × 10−25.88 × 10−22.58 × 10−51.54 × 10−53.63 × 10−42.10 × 10−4

Social impact analysis results demonstrate that lower manufacturing process times also lead to reduced nonfatal occupational injuries and illnesses, as well as reduced days away from work. Since the days away from work (DAW) metric is calculated based upon nonfatal occupational injuries and illnesses (NOI), a lower DAW is found for related parts. Despite the improved environmental performance of scenario SCa, from a social perspective SCb results in a reduced NOI and DAW for both manufacturing (14% each) and transportation (24% each). When considering the pedal assembly scenario, the best-performing assembly relies on a mix of SCa and SCb scenarios. The worst-performing assembly scenario has a carbon footprint of 3.5 times larger than the best-performing assembly scenario and causes 20% fewer nonfatal occupational injuries/illnesses and days away from work.

## Summary and Conclusions

The research problem explored herein is to better understand the potential impact of product manufacturing and supply chain activities on energy consumption and associated carbon footprint, as well as worker health (number of injuries/illnesses and associated days away from work) from a cradle-to-gate life cycle perspective. Energy use and carbon footprint have been identified as key issues, due to the associated impacts on energy availability and concerns over climate change. In addition, decisions within manufacturing industry directly impact labor conditions and worker health and safety issues. Providing a safer work environment for operators can reduce the risk of exposure to injuries and illnesses, and decrease their severity leading to reduced days away from work. Thus, several objectives were identified to address the research question posed. First, a comprehensive review composed largely of recent research literature has been reported to identify current research needs and support the novelty of this work. Second, parametric models are devised to compute process times and energy consumption, as well as the associated carbon footprint, nonfatal occupational injuries and illnesses, and days away from work.

Literature related to sustainable manufacturing and supply chain related research is reported from the 1990s to 2015. Most prior work is either theory-driven or centered on a particular process. In addition, most prior work is focused on evaluating environmental impacts, with some initial studies exploring social impacts from a conceptual perspective. Manufacturing process planning has been identified to be an area that is underevaluated. Similarly, supply chain research lacks network design planning from a sustainability performance perspective. Several studies concluded current research is deficient in linking energy use and GHG emissions impacts during supply chain operations. To address these limitations, a process-based modeling approach has been utilized to simultaneously evaluate unit manufacturing process and supply chain network design scenarios.

The approach uses carbon footprint as an environmental performance measure, based on analytical models of energy consumption, and the worker health metrics noted above as a social performance measures, based on calculated transportation and processing times and safety statistics reported by the U.S. Bureau of Labor Statistics. Ten manufacturing process models were implemented in an Excel spreadsheet tool, including injection molding, metal casting, metal cutting, forging, bending, and various machining processes. In addition, various transportation modes (e.g., road, rail, barge, and air freight), routes, and mass to be transported were considered in supply chain impact models.

Simultaneous consideration of manufacturing processes and supply chain network designs provides the opportunity to achieve better environmental and social performance for a given product and can yield to better understanding of the environmental and social impacts generated by the manufacturing processes and supply chain activities at the design stage. The manufacturing upstream processes are often the most impactful in the cradle-to-gate life cycle of the product as demonstrated for the pedal body plate and the other pedal components. Thus, it is imperative to utilize upstream processes that are not energy and carbon intensive (e.g., the EAF as opposed to the BOF in steelmaking). When air freight transport is employed, transportation becomes more impactful than manufacturing activities. Manufacturing process planning can be conducted in a synchronized manner with supply chain network design to replace impactful processes identified within the supply chain in pursuit of enhancing product sustainability performance across the life cycle. Global supply chains can utilize sea transport for the greatest possible reduction of carbon footprint. Local transportation is better utilized by either rail or intermodal rail/road travel, which can reduce carbon footprint by 58–65% over road travel alone.

The foregoing review furnishes a starting point for identifying future research needs and direction. Currently, for example, the framework is realized through scenario analysis. Tools could be developed by future work under this framework that incorporate mathematical optimization or system dynamics as approaches to identify the best manufacturing process and supply chain configurations. In this case, work will need to consider how to balance tradeoffs and conduct multi-objective optimization with respect to various aspects of sustainability. Moreover, the demonstrated application of the framework motivates researchers to continue to pursue the challenges of tradeoff analysis when assessing competing objectives simultaneously. This is especially apparent when faced with uncertain data and value-laden comparisons, as in the case of reduction of environmental and social impacts across the product manufacturing supply chain presented here.

## Acknowledgment

This material is based upon work supported by the National Science Foundation under Grant Nos. OCI-1041423, OCI-1041380, and OCI-1041328. The authors wish to thank Ms. Natalie Dupuy from Crescent Valley High School (Corvallis, OR) for her assistance with this research during her ASE internship.

## Nomenclature

• a =

bend angle

•
• D =

depth of cut

•
• dj =

•
• Df =

final diameter

•
• Di =

initial diameter

•
• D0 =

initial rod diameter

•
• DAWmfg =

days away from work for manufacturing

•
• DAWtran =

days away from work for transportation

•
• em(j) =

carbon equivalent emissions due to manufacturing process i at location j

•
• et(j) =

carbon equivalent emissions due to transportation from location j − 1 to location j

•
• E(i, j) =

energy consumption of manufacturing process i at location j

•
• ECD =

drilling energy consumption

•
• ECF =

forging energy consumption

•
• ECL =

laser energy consumption

•
• ECT =

turning energy consumption

•
• FB =

force required to bend

•
• FF =

forging force

•
• h =

rod height

•
• hf =

final rod height

•
• LT =

length of cut

•
• m =

mass to be transported

•
• MDAWmfg =

median days away from work for manufacturing

•
• MDAWtran =

median days away from work for transportation

•
• Ntran =

number of parts that can be transported by each vehicle

•
• NOImfg =

nonfatal occupational injuries for manufacturing

•
• NOItran =

nonfatal occupational injuries for transportation

•
• P =

laser power

•
• Pactual =

actual power

•
• PV =

production volume

•
• r =

•
• R =

•
• RDAWmfg =

days away from work rate for manufacturing

•
• RDAWtran =

days away from work rate for transportation

•
• RNOImfg =

nonfatal occupational injury rate for manufacturing

•
• RNOItran =

nonfatal occupational injury rate for transportation

•
• S =

material strength coefficient

•
• t =

plate thickness

•
• Tmfg =

manufacturing process time

•
• Ttran =

transportation time

•
• U =

unit power

•
• Wd =

die width

•
• X =

length of cut

•
• Y =

flow stress

•
• Yf =

flow stress of the material

•
• αe =

energy to carbon footprint conversion factor

•
• βm =

average emission factor of the utilized transportation mode

•
• εf =

true strain

•
• ηd =

drilling efficiency

•
• ηf =

forging efficiency

•
• ηl =

laser cutting efficiency

•
• ηt =

turning efficiency

•
• μ =

coefficient of friction

•
• νf =

velocity

•
• νl =

scanning velocity

•
• σT =

true stress

## References

1.
Brundtland
,
G. H.
,
1987
,
Our Common Future, World Commission on Environment and Development (WCED)
,
Oxford University Press
, Oxford, UK.
2.
Intergovernmental Panel on Climate Change (IPCC)
,
2007
,
Climate Change 2007—The Physical Science Basis: Working Group I Contribution to the Fourth Assessment Report of the IPCC
,
Cambridge University Press
, New York.
3.
Hunter
,
J.
,
2009
, “
Estimating Sea-Level Extremes Under Conditions of Uncertain Sea-Level Rise
,”
Clim. Change
,
99
(
3–4
), pp.
331
350
.
4.
Elabras Veiga
,
L. B.
, and
Magrini
,
A.
,
2009
, “
Eco-Industrial Park Development in Rio de Janeiro, Brazil: A Tool for Sustainable Development
,”
J. Cleaner Prod.
,
17
(
7
), pp.
653
661
.
5.
Kremer
,
G. E.
,
Haapala
,
K.
,
Murat
,
A.
,
Chinnam
,
R. B.
,
Kim
,
K.
,
Monplaisir
,
L.
, and
Lei
,
T.
,
2016
, “
Directions for Instilling Economic and Environmental Sustainability Across Product Supply Chains
,”
J. Cleaner Prod.
,
112
(
Pt. 3
), pp.
2066
2078
.
6.
Hutchins
,
M. J.
,
Gierke
,
J. S.
, and
Sutherland
,
J. W.
,
2009
, “
Decision Making for Social Sustainability: A Life-Cycle Assessment Approach
,”
IEEE International Symposium on Technology and Society
, ISTAS'09, pp.
1
5
.
7.
Seuring
,
S.
,
2013
, “
A Review of Modeling Approaches for Sustainable Supply Chain Management
,”
Decis. Support Syst.
,
54
(
4
), pp.
1513
1520
.
8.
Fuisz-Kehrbach
,
S.-K.
,
2015
, “
A Three-Dimensional Framework to Explore Corporate Sustainability Activities in the Mining Industry: Current Status and Challenges Ahead
,”
Resour. Policy
,
46
(
Pt. 1
), pp.
101
115
.
9.
Wirth
,
H.
,
Kulczycka
,
J.
,
Hausner
,
J.
, and
Koński
,
M.
,
2016
, “
Corporate Social Responsibility: Communication About Social and Environmental Disclosure by Large and Small Copper Mining Companies
,”
Resour. Policy
,
49
, pp.
53
60
.
10.
Gunasekaran
,
A.
, and
Spalanzani
,
A.
,
2012
, “
Sustainability of Manufacturing and Services: Investigations for Research and Applications
,”
Int. J. Prod. Econ.
,
140
(
1
), pp.
35
47
.
11.
Jørgensen
,
A.
,
Bocq
,
A.
,
Nazarkina
,
L.
, and
Hauschild
,
M.
,
2007
, “
Methodologies for Social Life Cycle Assessment
,”
Int. J. Life Cycle Assess.
,
13
(
2
), pp.
96
103
.
12.
Parris
,
T. M.
, and
Kates
,
R. W.
,
2003
, “
Characterizing and Measuring Sustainable Development
,”
Annu. Rev. Environ. Resour.
,
28
(
1
), pp.
559
586
.
13.
Golini
,
R.
,
Longoni
,
A.
, and
Cagliano
,
R.
,
2014
, “
Developing Sustainability in Global Manufacturing Networks: The Role of Site Competence on Sustainability Performance
,”
Int. J. Prod. Econ.
,
147
(
Pt. B
), pp.
448
459
.
14.
Yawar
,
S. A.
, and
Seuring
,
S.
, “
Management of Social Issues in Supply Chains: A Literature Review Exploring Social Issues, Actions and Performance Outcomes
,”
J. Bus. Ethics
, epub, pp. 1–23.
15.
Hutchins
,
M. J.
, and
Sutherland
,
J. W.
,
2008
, “
An Exploration of Measures of Social Sustainability and Their Application to Supply Chain Decisions
,”
J. Cleaner Prod.
,
16
(
15
), pp.
1688
1698
.
16.
Eastwood
,
M. D.
, and
Haapala
,
K. R.
,
2015
, “
A Unit Process Model Based Methodology to Assist Product Sustainability Assessment During Design for Manufacturing
,”
J. Cleaner Prod.
,
108
(
Pt. A
), pp.
54
64
.
17.
Wilhelm
,
M.
,
Hutchins
,
M.
,
Mars
,
C.
, and
Benoit-Norris
,
C.
,
2015
, “
An Overview of Social Impacts and Their Corresponding Improvement Implications: A Mobile Phone Case Study
,”
J. Cleaner Prod.
,
102
, pp.
302
315
.
18.
Zhang
,
H.
, and
Haapala
,
K. R.
,
2015
, “
Integrating Sustainable Manufacturing Assessment Into Decision Making for a Production Work Cell
,”
J. Cleaner Prod.
,
105
, pp.
52
63
.
19.
Székely
,
F.
, and
Knirsch
,
M.
,
2005
, “
Responsible Leadership and Corporate Social Responsibility: Metrics for Sustainable Performance
,”
Eur. Manage. J.
,
23
(
6
), pp.
628
647
.
20.
Ma
,
J.
, and
Okudan Kremer
,
G. E.
,
2015
, “
A Fuzzy Logic-Based Approach to Determine Product Component End-of-Life Option From the Views of Sustainability and Designer's Perception
,”
J. Cleaner Prod.
,
108
(
Pt. A
), pp.
289
300
.
21.
Chen
,
D.
,
Heyer
,
S.
,
Ibbotson
,
S.
,
Salonitis
,
K.
,
Steingrímsson
,
J. G.
, and
Thiede
,
S.
,
2015
, “
Direct Digital Manufacturing: Definition, Evolution, and Sustainability Implications
,”
J. Cleaner Prod.
,
107
, pp.
615
625
.
22.
Sancha
,
C.
,
Gimenez
,
C.
, and
Sierra
,
V.
,
2016
, “
Achieving a Socially Responsible Supply Chain Through Assessment and Collaboration
,”
J. Cleaner Prod.
,
112
(
Pt. 3
), pp.
1934
1947
.
23.
U.S. EPA,
2011
, “
How Clean Is the Electricity I Use?—Power Profiler | Clean Energy | U.S. EPA
,” U.S. Environmental Protection Agency, Washington, DC, last accessed Feb. 8, 2015, http://www.epa.gov/cleanenergy/energy-and-you/how-clean.html
24.
Sutherland
,
J. W.
, and
Gunter
,
K. L.
,
2001
, “
Environmental Attributes of Manufacturing Processes
,”
Handbook of Environmentally Conscious Manufacturing
,
Springer
, Berlin, pp.
293
316
.
25.
Jeswiet
,
J.
, and
Hauschild
,
M.
,
2005
, “
EcoDesign and Future Environmental Impacts
,”
Mater. Des.
,
26
(
7
), pp.
629
634
.
26.
Ramani
,
K.
,
Ramanujan
,
D.
,
Bernstein
,
W. Z.
,
Zhao
,
F.
,
Sutherland
,
J.
,
Handwerker
,
C.
,
Choi
,
J.-K.
,
Kim
,
H.
, and
Thurston
,
D.
,
2010
, “
Integrated Sustainable Life Cycle Design: A Review
,”
ASME J. Mech. Des.
,
132
(
9
), p.
091004
.
27.
Zhang
,
W.
,
2011
,
Intelligent Energy Field Manufacturing: Interdisciplinary Process Innovations
,
CRC Press
, Boca Raton, FL.
28.
Sutherland
,
J. W.
,
Gunter
,
K. L.
,
Haapala
,
K. R.
,
,
K.
,
Skerlos
,
S. J.
,
Zimmerman
,
J. B.
,
Olson
,
W. W.
, and
,
R.
,
2003
, “
Environmentally Benign Manufacturing: Status and Vision for the Future
,”
31st North American Manufacturing Research Conference
, Hamilton, ON, Canada, May 20–23, Vol.
31
, pp.
345
352
.
29.
Duflou
,
J. R.
,
Sutherland
,
J. W.
,
Dornfeld
,
D.
,
Herrmann
,
C.
,
Jeswiet
,
J.
,
Kara
,
S.
,
Hauschild
,
M.
, and
Kellens
,
K.
,
2012
, “
Towards Energy and Resource Efficient Manufacturing: A Processes and Systems Approach
,”
CIRP Ann.-Manuf. Technol.
,
61
(
2
), pp.
587
609
.
30.
Gutowski
,
T. G.
,
Murphy
,
C. F.
,
Allen
,
D. T.
,
Bauer
,
D. J.
,
Bras
,
B.
,
Piwonka
,
T. S.
,
Sheng
,
P. S.
,
Sutherland
,
J. W.
,
Thurston
,
D. L.
, and
Wolff
,
E. E.
,
2001
, “
Environmentally Benign Manufacturing
,” World Technology (
WTEC
) Division, International Technology Research Institute, Baltimore, MD, epub.
31.
Fiksel
,
J.
,
1993
, “
Design for Environment: An Integrated Systems Approach
,”
IEEE International Symposium on Electronics and the Environment
, Arlington, VA, May 10–12, pp.
126
131
.
32.
Chiu
,
M.-C.
,
Alsaffar
,
A. J.
,
Okudan
,
G. E.
, and
Haapala
,
K. R.
,
2010
, “
Reducing Supply Chain Costs and Carbon Footprint During Product Design
,”
IEEE International Symposium on Sustainable Systems and Technology (ISSST)
, Washington, DC, May 17–19, pp.
1
6
.
33.
Fullan
,
M.
,
2005
,
Leadership & Sustainability: System Thinkers in Action
,
Corwin Press
,
Thousand Oaks, CA
.
34.
Hargreaves
,
A.
, and
Fink
,
D.
,
2005
,
, Vol.
6
,
Jossey-Bass
,
San Francisco, CA
.
35.
U.S. EPA,
2007
, “
EPA's Emissions & Generation Resource Integrated Database (eGRID): Power Profiler eGRID Subregion and GHG Emissions Finder Tool
,” U.S. Environmental Protection Agency, Washington, DC.
36.
Hertwich
,
E. G.
,
Hammitt
,
J. K.
, and
Pease
,
W. S.
,
2000
, “
A Theoretical Foundation for Life‐Cycle Assessment
,”
J. Ind. Ecol.
,
4
(
1
), pp.
13
28
.
37.
Corbett
,
C. J.
, and
Kirsch
,
D. A.
,
2001
, “
International Dispersion of ISO 14000 Certifications
,”
Prod. Oper. Manage.
,
10
(
3
), pp.
327
342
.
38.
Giachetti
,
R. E.
,
1998
, “
A Decision Support System for Material and Manufacturing Process Selection
,”
J. Intell. Manuf.
,
9
(
3
), pp.
265
276
.
39.
Lin
,
X.
, and
Polenske
,
K. R.
,
1998
, “
Input–Output Modeling of Production Processes for Business Management
,”
Struct. Change Econ. Dyn.
,
9
(
2
), pp.
205
226
.
40.
Hernandez-Matias
,
J. C.
,
Vizan
,
A.
,
Hidalgo
,
A.
, and
Rios
,
J.
,
2006
, “
Evaluation of Techniques for Manufacturing Process Analysis
,”
J. Intell. Manuf.
,
17
(
5
), pp.
571
583
.
41.
Rajemi
,
M. F.
,
Mativenga
,
P. T.
, and
Aramcharoen
,
A.
,
2010
, “
Sustainable Machining: Selection of Optimum Turning Conditions Based on Minimum Energy Considerations
,”
J. Cleaner Prod.
,
18
(
10
), pp.
1059
1065
.
42.
ISO
,
2014
, “
Machine Tools—Environmental Evaluation of Machine Tools—Part 1: Design Methodology for Energy-Efficient Machine Tools
,” International Organization for Standardization, Geneva, Switzerland, Standard No. ISO 14955-1:2014.
43.
Diaz
,
N.
,
Redelsheimer
,
E.
, and
Dornfeld
,
D.
,
2011
, “
Energy Consumption Characterization and Reduction Strategies for Milling Machine Tool Use
,”
Glocalized Solutions for Sustainability in Manufacturing
,
J.
Hesselbach
, and
Herrmann
,
C.
, eds.,
Springer
,
Berlin
, pp.
263
267
.
44.
Gutowski
,
T.
,
Dahmus
,
J.
, and
Thiriez
,
A.
,
2006
, “
Electrical Energy Requirements for Manufacturing Processes
,”
13th CIRP International Conference on Life Cycle Engineering
, Leuven, Belgium, May 31–June 2, Vol.
5
, pp.
560
564
.
45.
Haapala
,
K. R.
,
,
K. N.
, and
Sutherland
,
J. W.
,
2004
, “
Predicting Manufacturing Waste and Energy for Sustainable Product Development Via WE-Fab Software
,”
Global Conference on Sustainable Product Development and Life Cycle Engineering
, Berlin, Germany, Sept. 29–Oct. 1, pp.
243–
250
.
46.
,
B.
,
Das
,
J.
,
Lam
,
M.
, and
Ly
,
C.
,
2014
, “
Sustainability Indicators for Finishing Operations Based on Process Performance and Part Quality
,”
Procedia CIRP
,
14
, pp.
564
569
.
47.
,
J.
,
Mani
,
M.
,
Lee
,
J. H.
, and
Lyons
,
K. W.
,
2015
, “
Energy Performance Evaluation and Improvement of Unit-Manufacturing Processes: Injection Molding Case Study
,”
J. Cleaner Prod.
,
105
, pp.
157
170
.
48.
Kellens
,
K.
,
Dewulf
,
W.
,
Overcash
,
M.
,
Hauschild
,
M. Z.
, and
Duflou
,
J. R.
,
2012
, “
Methodology for Systematic Analysis and Improvement of Manufacturing Unit Process Life Cycle Inventory (UPLCI) CO2PE! Initiative (Cooperative Effort on Process Emissions in Manufacturing). Part 1: Methodology Description
,”
Int. J. Life Cycle Assess.
,
17
(
1
), pp.
69
78
.
49.
Kellens
,
K.
,
Dewulf
,
W.
,
Overcash
,
M.
,
Hauschild
,
M. Z.
, and
Duflou
,
J. R.
,
2012
, “
Methodology for Systematic Analysis and Improvement of Manufacturing Unit Process Life Cycle Inventory (UPLCI) CO2PE! Initiative (Cooperative Effort on Process Emissions in Manufacturing). Part 2: Case Studies
,”
Int. J. Life Cycle Assess.
,
17
(
2
), pp.
242
251
.
50.
Overcash
,
M.
,
Twomey
,
J.
, and
Kalla
,
D.
,
2009
, “
Unit Process Life Cycle Inventory for Product Manufacturing Operations
,”
ASME
Paper No. MSEC2009-84065.
51.
Branker
,
K.
,
Jeswiet
,
J.
, and
Kim
,
I. Y.
,
2011
, “
Greenhouse Gases Emitted in Manufacturing a Product—A New Economic Model
,”
CIRP Ann.-Manuf. Technol.
,
60
(
1
), pp.
53
56
.
52.
Nava
,
P.
,
2009
, “
Minimizing Carbon Emissions in Metal Forming
,” Master's thesis, Queens University, Kingston, ON, Canada.
53.
Jeswiet
,
J.
, and
Nava
,
P.
,
2009
, “
Applying CES to Assembly and Comparing Carbon Footprints
,”
Int. J. Sustainable Eng.
,
2
(
4
), pp.
232
240
.
54.
Laurent
,
A.
,
Olsen
,
S. I.
, and
Hauschild
,
M. Z.
,
2010
, “
Carbon Footprint as Environmental Performance Indicator for the Manufacturing Industry
,”
CIRP Ann.-Manuf. Technol.
,
59
(
1
), pp.
37
40
.
55.
Fang
,
K.
,
Uhan
,
N.
,
Zhao
,
F.
, and
Sutherland
,
J. W.
,
2011
, “
A New Approach to Scheduling in Manufacturing for Power Consumption and Carbon Footprint Reduction
,”
J. Manuf. Syst.
,
30
(
4
), pp.
234
240
.
56.
Dietmair
,
A.
, and
Verl
,
A.
,
2008
, “
Energy Consumption Modeling and Optimization for Production Machines
,”
IEEE International Conference on Sustainable Energy Technologies
, ICSET 2008, pp.
574
579
.
57.
Sun
,
Z.
, and
Li
,
L.
,
2014
, “
Potential Capability Estimation for Real Time Electricity Demand Response of Sustainable Manufacturing Systems Using Markov Decision Process
,”
J. Cleaner Prod.
,
65
, pp.
184
193
.
58.
Sun
,
Z.
,
Biller
,
S.
,
Gu
,
F.
, and
Li
,
L.
,
2011
, “
Energy Consumption Reduction for Sustainable Manufacturing Systems Considering Machines With Multiple-Power States
,”
ASME
Paper No. MSEC2011-50069.
59.
Ilgin
,
M. A.
, and
Gupta
,
S. M.
,
2010
, “
Environmentally Conscious Manufacturing and Product Recovery (ECMPRO): A Review of the State of the Art
,”
J. Environ. Manage.
,
91
(
3
), pp.
563
591
.
60.
Seuring
,
S.
, and
Müller
,
M.
,
2008
, “
From a Literature Review to a Conceptual Framework for Sustainable Supply Chain Management
,”
J. Cleaner Prod.
,
16
(
15
), pp.
1699
1710
.
61.
Cholette
,
S.
, and
Venkat
,
K.
,
2009
, “
The Energy and Carbon Intensity of Wine Distribution: A Study of Logistical Options for Delivering Wine to Consumers
,”
J. Cleaner Prod.
,
17
(
16
), pp.
1401
1413
.
62.
Sundarakani
,
B.
,
De Souza
,
R.
,
Goh
,
M.
,
Wagner
,
S. M.
, and
Manikandan
,
S.
,
2010
, “
Modeling Carbon Footprints Across the Supply Chain
,”
Int. J. Prod. Econ.
,
128
(
1
), pp.
43
50
.
63.
Saunders
,
C.
, and
Barber
,
A.
,
2007
,
Comparative Energy and Greenhouse Gas Emissions of New Zealand's and the UK's Dairy Industry
,
Agribusiness and Economics Research Unit, Lincoln University
,
Lincoln, New Zealand
, Report No. 297.
64.
Ibbotson
,
S.
, and
Kara
,
S.
,
2011
, “
Carbon footprint Analysis of Products Cradle-to-Gate and Manufacturing Supply Chains
,”
7th Australian Conference on Life Cycle Assessment
, East Melbourne, Australia, Mar. 9–10, pp.
1
10
.
65.
Herrmann
,
C.
,
Kara
,
S.
,
Thiede
,
S.
, and
Luger
,
T.
,
2010
, “
Energy Efficiency in Manufacturing—Perspectives From Australia and Europe
,”
CIRP International Conference on Life Cycle Engineering
, pp.
23
28
.
66.
Arnette
,
A. N.
,
Brewer
,
B. L.
, and
Choal
,
T.
,
2014
, “
Design for Sustainability (DFS): The Intersection of Supply Chain and Environment
,”
J. Cleaner Prod.
,
83
, pp.
374
390
.
67.
Bevilacqua
,
M.
,
Ciarapica
,
F. E.
,
Giacchetta
,
G.
, and
Marchetti
,
B.
,
2011
, “
A Carbon Footprint Analysis in the Textile Supply Chain
,”
Int. J. Sustainable Eng.
,
4
(
1
), pp.
24
36
.
68.
Govindan
,
K.
,
Azevedo
,
S. G.
,
Carvalho
,
H.
, and
,
V.
,
2014
, “
Impact of Supply Chain Management Practices on Sustainability
,”
J. Cleaner Prod.
,
85
, pp.
212
225
.
69.
Ahi
,
P.
, and
Searcy
,
C.
,
2015
, “
Assessing Sustainability in the Supply Chain: A Triple Bottom Line Approach
,”
Appl. Math. Modell.
,
39
(
10–11
), pp.
2882
2896
.
70.
Garcia
,
D. J.
, and
You
,
F.
,
2015
, “
Supply Chain Design and Optimization: Challenges and Opportunities
,”
Comput. Chem. Eng.
,
81
, pp.
153
170
.
71.
Sheng
,
P.
, and
Worhach
,
P.
,
1998
, “
A Process Chaining Approach Toward Product Design for Environment
,”
J. Ind. Ecol.
,
1
(
4
), pp.
35
55
.
72.
Reap
,
J.
,
Román
,
F.
,
Guldberg
,
T.
, and
Bras
,
B.
,
2006
, “
Integrated Ecosystem Landscape and Industrial Modeling for Strategic Environmentally Conscious Process Technology Selection
,”
13th CIRP International Conference on Life Cycle Engineering
, Leuven, Belgium, May 31–June 2, pp.
213
218
.
73.
De Benedetto
,
L.
, and
Klemeš
,
J.
,
2009
, “
The Environmental Performance Strategy Map: An Integrated LCA Approach to Support the Strategic Decision-Making Process
,”
J. Cleaner Prod.
,
17
(
10
), pp.
900
906
.
74.
Bright
,
R. M.
,
Cherubini
,
F.
, and
Strømman
,
A. H.
,
2011
, “
Climate Impacts of Bioenergy: Inclusion of Carbon Cycle and Albedo Dynamics in Life Cycle Impact Assessment
,”
Environ. Impact Assess. Rev.
,
37
, pp.
2
11
.
75.
Haapala
,
K. R.
,
Zhao
,
F.
,
Camelio
,
J.
,
Sutherland
,
J. W.
,
Skerlos
,
S. J.
,
Dornfeld
,
D. A.
,
Jawahir
,
I. S.
,
Clarens
,
A. F.
, and
Rickli
,
J. L.
,
2013
, “
A Review of Engineering Research in Sustainable Manufacturing
,”
ASME J. Manuf. Sci. Eng.
,
135
(
4
), p.
041013
.
76.
Blom
,
M.
, and
Solmar
,
C.
,
2009
, “
How to Socially Assess Biofuels: Case Study of the UNEP/SETAC Code of Practice for Social-Economical LCA
,” Master's thesis, Luleå University of Technology, Luleå, Sweden.
77.
Dreyer
,
L. C.
,
Hauschild
,
M. Z.
, and
Schierbeck
,
J.
,
2010
, “
Characterisation of Social Impacts in LCA: Part 1: Development of Indicators for Labour Rights
,”
Int. J. Life Cycle Assess.
,
15
(
3
), pp.
247
259
.
78.
BLS
,
2015
, “
Bureau of Labor Statistics
,” United States Department of Labor, Washington, DC, last accessed Dec. 10, 2015, http://www.bls.gov/
79.
U.S. EIA,
2011
, “
Energy Perspectives, Total Energy—Energy Information Administration
,” U.S. Energy Information Administration, Washington, DC, last accessed Feb. 8, 2015, http://www.eia.gov/totalenergy/data/annual/perspectives.cfm
80.
U.S. EIA,
2004
, “
Greenhouse Gases, Climate Change, and Energy
,” U.S. Energy Information Administration, Washington, DC, last accessed Feb. 8, 2012, http://www.eia.gov/oiaf/1605/ggccebro/chapter1.html
81.
U.S. NASA,
2015
, “
Climate Change: Vital Signs of the Planet: Causes
,” NASA, Washington, DC, Last accessed Feb. 8, 2015, http://climate.nasa.gov/causes/
82.
Wiedmann
,
T.
, and
Minx
,
J.
,
2008
, “
A Definition of ‘Carbon Footprint’
,”
Ecological Economics Research Trends
,
Nova Science Publishers
,
Hauppauge, NY
, pp.
1
11
.
83.
Wiedmann
,
T.
, and
Minx
,
J.
,
2007
, “
A Definition of ‘Carbon Footprint’
,”
Ecological Economics Research Trends
, Vol.
1
,
Nova Science Publishers
,
Hauppauge, NY
, pp.
1
11
.
84.
Boguski
,
T. K.
,
2010
, “
Life Cycle Carbon Footprint of the National Geographic Magazine
,”
Int. J. Life Cycle Assess.
,
15
(
7
), pp.
635
643
.
85.
Joyce
,
T.
,
Okrasinski
,
T. A.
, and
Schaeffer
,
W.
,
2010
, “
Estimating the Carbon Footprint of Telecommunications Products: A Heuristic Approach
,”
ASME J. Mech. Des.
,
132
(
9
), p.
094502
.
86.
U.S. National Research Council
,
1995
,
Unit Manufacturing Processes: Issues and Opportunities in Research
,
,
Washington, DC
.
87.
Kondoh
,
K.
, and
Aizawa
,
T.
,
2003
, “
Environmentally Benign Fabricating Process of Magnesium Alloy by Cyclical Plastic Working in Solid-State
,”
Mater. Trans.
,
44
(
7
), pp.
1276
1283
.
88.
Schifo
,
J. F.
, and
,
J. T.
,
2004
,
Theoretical/Best Practice Energy Use in Metalcasting Operations
, U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, Washington, DC.
89.
Ghany
,
K. A.
, and
Newishy
,
M.
,
2005
, “
Cutting of 1.2 mm Think Austenitic Stainless Steel Sheet Using Pulsed and CW Nd:YAG Laser
,”
J. Mater. Process. Technol.
,
168
(
3
), pp.
438
447
.
90.
Kordonowy
,
D. N.
,
2002
, “
A Power Assessment of Machining Tools
,” B.S. thesis, Massachusetts Institute of Technology, Cambridge, MA.
91.
Santos
,
J. P.
,
Oliveira
,
M.
,
Almeida
,
F. G.
,
Pereira
,
J. P.
, and
Reis
,
A.
,
2011
, “
Improving the Environmental Performance of Machine-Tools: Influence of Technology and Throughput on the Electrical Energy Consumption of a Press-Brake
,”
J. Cleaner Prod.
,
19
(
4
), pp.
356
364
.
92.
Quiza Sardiñas
,
R.
,
Rivas Santana
,
M.
, and
Alfonso Brindis
,
E.
,
2006
, “
Genetic Algorithm-Based Multi-Objective Optimization of Cutting Parameters in Turning Processes
,”
Eng. Appl. Artif. Intell.
,
19
(
2
), pp.
127
133
.
93.
Suda
,
S.
,
Yokota
,
H.
,
Inasaki
,
I.
, and
Wakabayashi
,
T.
,
2002
, “
A Synthetic Ester as an Optimal Cutting Fluid for Minimal Quantity Lubrication Machining
,”
CIRP Ann.-Manuf. Technol.
,
51
(
1
), pp.
95
98
.
94.
Kalpakjian
,
S.
,
2010
,
Manufacturing Engineering and Technology
,
Prentice Hall
,
New York, NY
.
95.
,
I.
, and
Hancock
,
G. F.
,
1979
,
Handbook of Industrial Energy Analysis
,
Halsted Press
,
Chichester, UK
.
96.
Alsaffar
,
A.
,
2012
, “
Process-Based Modeling for Cradle-to-Gate Energy and Carbon Footprint Reduction in Product Design
,” M.S. thesis, Oregon State University, Corvallis, OR.
97.
AK Steel,
2013
, “
316/316L Stainless Steel
,” AK Steel Corporation, West Chester, OH.
98.
Lange
,
K.
,
1985
,
Handbook of Metal Forming
,
McGraw-Hill
, New York.
99.
Kutz
,
M.
,
2008
,
Environmentally Conscious Transportation
,
Wiley, New York
.
100.
U.S. EPA,
2009
, “
eGRID2012 Version 1.0 Year 2009 Summary Tables
,” U.S. Environmental Protection Agency, Washington, DC.
101.
U.S. EIA,
2010
, “
International Electricity Emission Factors by Country
,” U.S. Energy Information Administration, Washington, DC.
102.
McKinnon
,
A. C.
,
2003
, “
Logistics and the Environment
,”
Handbook of Transport and the Environment
, Vol.
4
,
D. A.
Hensher
, and
K. J.
Button
, eds.,
Elsevier
,
Oxford, UK
, pp.
665
685
.
103.
104.
NASA
,
2015
, “
NASA–NSSDCA: Terrestrial Planet Mileage Calculator
,” NASA, Washington, DC, last accessed Jan. 16, 2016, http://nssdc.gsfc.nasa.gov/special/
105.
Stubbles
,
J.
,
2000
,
Energy Use in the U.S. Steel Industry: An Historical Perspective and Future Opportunities
, Energetics, Inc. and U.S. Department of Energy Office of Industrial Technologies, Washington, DC.
106.
Price
,
L.
,
Sinton
,
J.
,
Worrell
,
E.
,
Phylipsen
,
D.
,
Xiulian
,
H.
, and
Ji
,
L.
,
2002
, “
Energy Use and Carbon Dioxide Emissions From Steel Production in China
,”
Energy
,
27
(
5
), pp.
429
446
.