Industry, Commerce and Mining Energy Demands

It is a well-known and undoubted fact that heavy industries such as the steel and cement sectors, the buying and selling of goods and services and the mining industry are all large contributors to anthropogenic carbon pollution (Van Ruijven et al. 2016).

Several modeling studies to estimate energy consumption, modern use and the effects of emissions on the environment have been conducted, each model typically being sector-specific and tailored to variations and improvements on production practices and technology advances in the sector. One such modeling tool is the Clean Energy Manufacturing Analysis Center (CEMAC) materials flow through industry (MFI) supply chain modeling tool. The CEMAC MFI tool tracks materials, resources, energy, water use and GHG emissions throughout commodity supply chains (CEMAC 2016). The CEMAC MFI tool is based on a database of over 650 industrial commodities and close to 1400 input-output models of material and fuel consumption by manufacturing processes. The vast number of combinations and the database of gathered information stress the complexity of modeling an economically viable and environmentally friendly industry and implementing eco-planning for new and current developments. No single model exists that accounts for all variables in all industries, but computer-aided models provide a framework to develop application-specific models.

It is necessary to analyze and record trends occurring over the long term, typically several years, to account for seasonal changes in energy use, generated carbon emissions and potential modification strategies and trade-offs/limitations for each strategy. In construction projects, for example, the workflow of the construction life cycle is identified by first defining the goal and scope of the project, then by implementing a life cycle inventory and finally by providing a detailed impact assessment. Each major stage or subsystem of the overall system can be individually characterized and quantized and the environmental impact can be determined (Li and Chen 2016). Construction projects are not only scrutinized during development, but also at their completion because they play an important role in eco-city and sustainable development. Environmental modeling and carbon emission contribution should be considered during the construction phase and based on long-term operation through sustainable practice.

Developed models are typically bottom-up approaches, which are based on material flows, macro-economic models or econometric models, which account for socioeconomic indicators and material demand energy use and emissions (Van Ruijven et al. 2016). According to Van Ruijven et al. (2016), the best models to represent patterns in historical data are linearized regression models that relate economic activity as gross domestic product per capita to material consumption per capita. Non-linear models are used by Wang and Ye (2016), who introduce a power exponent into the traditional multivariate grey model (Julong 1989) to describe the non-linear relationship between carbon emissions from fossil fuel energy consumption and economic growth. grey systems are typically used to describe systems that lack structure, operation mechanism and behavior documenting (Julong 1989). These systems have been applied effectively to various disciplines such as agriculture, ecology, economy, meteorology, medicine, history, geography, industry, geology, hydrology, irrigation, the military, traffic management, material science, the environment, biological protection and the judicial system, with many more applications applying the theory of grey systems. A complete description of the grey system is provided by Julong (1989).

Linear regression models can be used to predict variable behavior, forecast changes in processes and technology improvements or focus on error reduction by fitting a predictive model to an observed and recorded dataset. Assumptions of linear regression models, which may affect the overall accuracy and feasibility of the results, are their weak exogenous behavior, restriction of forced linearity, constant variance across response variables, the assumption of uncorrelated errors and the lack of multi-collinearity in the predictor variables. A typical linear regression model, given a dataset

where y_t is the dependent variable on the linear p-vector of exogenous variable x_t. An unobserved random error variable e_t is used to characterize the discrepancy from the linear relationships between the dependent variable and the exogenous variable. The linear regression model used to forecast economic behavior of the material consumption process in industry in its simplest form is given by

where b is a p-dimensional parameter vector representing regression coefficients for statistical estimation and interference.

In general, models are based on case studies in specific industries (again, the most typical studies are conducted in the cement and steel manufacturing industries) with geographic specificity. Many studies have been conducted on industries in rapidly growing and expanding counties such as China, Thailand, Japan and India. Rootzen and Johnson (2016) examine the impacts of carbon pricing and investments in carbon abatement in the steel industry. A case study of passenger vehicle manufacturing and the effects of steel price increases on cost structures, as well as price variations at each step in the supply chain, was assessed. Rootzen and Johnson (2016) conclude that in order to achieve a return on investment in low-carbon steelmaking processes, specific to this case study, the price of steel and inevitably the manufactured end-user products will increase dramatically, primarily owing to the large capital investment and long-term maintenance. This trend is unfortunately witnessed throughout industry, especially in established and long-running factories, mines, power plants and commerce. The trend is also seen on a lower scale, such as in households where initial capital investment in low-carbon practices still outweighs the viability of return on investment. Eco-city modeling is therefore only as effective as the potential for its uptake in industry and the commitments from several industries or individuals to absorb initial capital investments.

Gouldson et al. 2015 analyze the urban action scenario developed by Erickson and Tempest (2014) and provide a specific methodology to approximate superfluous costs as well as the benefits of exploring the potential contributions of cities in the global scenario. This methodology analysis offers useful parameters to estimate the economic impact of efficient urbanization. These parameters concern commercial and residential buildings, passenger and freight transport and waste management. To model a low-carbon city, it is important to include quantifiable parameters, which provide a positive or negative weighting factor to improvements and energy-efficient planning. In the commercial and residential building environment, concerning either new buildings or energy-efficient retrofitting of older, existing buildings, four crucial datasets should be accounted for. The economic impact of efficient buildings is determined by

• • determining the cost of new buildings or retrofitting older buildings with low-carbon appliances, strategic lighting, temperature control and low-carbon materials; compared to an estimated (average) baseline for building. These baselines are commonly available based on the area where the development is planned and the type of building materials, specified in the local currency per m²;
• • secondly, the time associated with new energy-efficient developments or retrofitting of older buildings as a parameter of development area per year. This parameter can also be specified in development area per month, depending on the predicted time of completion;
• • quantifying the economic impact, annual savings in energy consumption (specified in kWh per year), compared to a baseline of similar construction without additional energy-savings strategies, which provides a useful and quantifiable parameter to measure the long-term sustainability and value-added benefits of these developments; and
• • finally, as specified by Gouldson et al. (2015), specifying the quantity of non-renewable fuel usage avoided through efficient developments as a percentage of the total energy consumption in a building.

These parameters are valuable contributions to determine the viability of additional expenditure on energy saving compared to the potential benefits and cost savings over the lifetime of the development or to determine the future estimated break-even point where the initial investment is superseded by annual savings in energy. Also covered in Gouldson et al. (2015) and listed under commercial and residential buildings are economic variables of appliances and lighting concerning the cost of saving a unit of electricity, specified in local currency per kWh, annual electricity savings in kWh per year and the energy savings per efficient appliance. The global drive towards photovoltaic cells to replace or contribute to energy generation also requires quantifiable parameters such as energy capacity factors, installation and maintenance cost and annual electricity savings compared to a traditional energy-generation baseline.

The following section reviews the modeling requirements and constraints used to estimate transport-based carbon emissions from various scenarios, fuel types and vehicle types.