Efficiency stagnation in global steel production urges joint supply- and demand-side mitigation efforts

Steel production is a difficult-to-mitigate sector that challenges climate mitigation commitments. Efforts for future decarbonization can benefit from understanding its progress to date. Here we report on greenhouse gas emissions from global steel production over the past century (1900-2015) by combining material flow analysis and life cycle assessment. We find that ~45 Gt steel was produced in this period leading to emissions of ~147 Gt CO2-eq. Significant improvement in process efficiency (~67%) was achieved, but was offset by a 44-fold increase in annual steel production, resulting in a 17-fold net increase in annual emissions. Despite some regional technical improvements, the industry’s decarbonization progress at the global scale has largely stagnated since 1995 mainly due to expanded production in emerging countries with high carbon intensity. Our analysis of future scenarios indicates that the expected demand expansion in these countries may jeopardize steel industry’s prospects for following 1.5 °C emission reduction pathways. To achieve the Paris climate goals, there is an urgent need for rapid implementation of joint supply- and demand-side mitigation measures around the world in consideration of regional conditions.


S1. Overview
Our analysis is performed based on the procedures given in Fig.S1.1, which include three major steps. In Section S1, we first clarify our system boundary, and explore the historical development of the studied processes (and technologies) through the compressive investigation of various technical reports and publications. Secondly, in the entire Section S2, we present detailed procedures for our material flow analysis and environmental impact analysis. This allows for generating results related to material stocks and flows throughout each studied process along the global steel cycle from 1900 to 2015, as well as the trend in greenhouse gas (GHG) emission of each studied process during 1900-2015. By combining these two sets of results, we obtain the historical trend of total GHG emissions from global steel production in the past 115 years, and perform uncertainty analysis of the results (the method is described in Section S2.3 and the results are presented in Section S3). Furthermore, in Step 3 (to enrich the implications of our results), we further deepen our analysis of the global level by looking at regional aspects. This is done by separating the World into 8 regions (i.e. Europe, North America, Developed Asia and Oceania, China, India, Developing Asia and Middle East, Latin America and Caribbean, Africa). Here, we further quantify the regional material stocks and flows since 1995, and examine their GHG emission intensity performance. Meanwhile, we also investigate different types of low-carbon technologies related to steel production and perform a scenario analysis until 2050 to provide recommendations on the required mitigation strategies for achieving climate change targets.

S1.1 System description
The global steel life cycle system is constructed in Fig. S1.2, which consists of 5 main stages (i.e. Mining, Material Preparation, Ironmaking, Steelmaking, and Steel finishing) in the material production system (as the main scope for further emission calculation), and the remaining three life cycle stages (i.e. Fabrication, Inuse and End-of-life). The main features of this structure are summarized as follows: Firstly, the material production stages from mining to steelmaking are constructed based on the guidance from World steel association 1 . Furthermore, The linkage from steelmaking, finishing, to fabrication is constructed based on 2 .
Thirdly, this study follows the treatment in 3    Oxygen furnace) have been applied during the period from 1900 to 2015. The historical share of each technology can be found in Fig. S1.4 20 . For the open-hearth method, Puddling process was first used to produce puddle iron and mild steel 21 .
Meanwhile, the Crucible is one of the oldest processes for melting purpose and fed with wrought iron and coke 22 . Then the open-hearth furnaces (OHF) (i.e. Siemens-Martin process, appeared in 1865) gradually replaced the Puddling process and dominated the entire steelmaking stage 23 . For convert steelmaking, the Bessemer process was one of the most fuel-saving innovations to make inexpensive steel in the early period 24 . However, the Bessemer (and Thomas) process had a bad flexibility and control in steel quality 25 , which was gradually replaced by open hearth technology. Then the development of Basic oxygen furnaces (BOF) in convert steelmaking become to dominate the steel making process due to the development of a method to separate oxygen from nitrogen on an industrial scale since the 1960s 26 . For the scrap-based technology, Electric arc furnace (EAF) is the major technique to produce steel at present, which was firstly introduced in the late nineteenth century 10 . Accordingly, the application period of Pudding, Bessemer and Thomas, OHF, BOF, Crucible, and EAF is assumed to be 1900-1925, 1900-1975, 1900-2015, 1955-2015, 1900-1921, and 1900-2015, respectively. Iron Foundry (IF) is assumed as one stage to include a series of process from iron melting, molding, casting, etc. to produce the cast iron products. It has a long history and still is applied at present 2 . Hence, its application period is assumed to be 1900-2015 in this study. S8 (7) Steel casting process (period: 1955-2015 for Continuous casting, and 1900-2015 for Ingot casting) Steel castings are used to deliver required strength or shock resistance to crude steel for subsequent rolling in the finishing mills. There are two main techniques in this stage: ingot casting and continuous casting (with liquid steel for castings). Continuous casting (CC) was introduced in the 1950s and has occupied this stage by 96.2% in 2015. Before the introduction of Continuous casting, the non-continuous casting was the primary technology to produce the cast steel product 27 . The division of these two techniques from 1955 to 2015 is obtained from 27 and world steel yearbook 28 . The steel finishing stage is to produce final steel products with various required shapes and properties, which includes a series of process like reheating, forming, shaping, drilling, welding, galvanizing. The rolling and finishing processes have been applied for a long history. For instance, the cold and hot rolling processes can be traced back to the 14 th and 17 th century, respectively. Hence, the application periods of those processes are all assumed to be 1900-2015 in this study. Due to the lack of statistical data regarding the production from different processes, this study assumes the history of finishing process would follow the contemporary structure as mapped in 2 . Basically, the steel from ingot casting would enter hot rolling mills or become the cast steel for direct use. Meanwhile, the steel from continuous casting would then enter hot rolling mills for reheating and rolling. The hot rolling mills include Section mill, Rod and Bar mill, Plate mill, and Hot strip mill. For simplicity, the cold rolling and finishing stage is treated as one process to produce final cold rolling (CR) products in this study.

S2. Quantitative Method
This section gives the detailed step for the material flow analysis and its associated emission calculation.

S2.1 Material Flow Quantification
This part aims to quantify the stocks, flows, losses within an anthropogenic cycle of steel annually from 1900 to 2015. The part begins with the anthropogenic cycle construction, which follows the basic life cycle stages (i.e. production, manufacturing, in-use, and end-of-life). Several sub-stages are given for each stage.
Especially, in the production stage, more detailed sub-stages are given according to the production. For the rest three stages (i.e. manufacturing, in-use and end-of-life stages), their stocks, inflows, and outflows can be obtained based on four major groups of steel products (i.e. construction, transportation, machinery, and durable daily goods). Based on the method of dynamic material flow, the inflow and outflow for each production technology are obtained, and the entire anthropogenic cycle of steel can be quantified.
(1) Production activities data sets As shown in Figure S2.1, there are 22 processes (in the blocks) in the material production system. Material flow analysis is applied as the primary approach to obtain the mass inflow, outflow, or loss for each process, which all are converted to iron (ferrous) content based on the ratio in the reference 29 . The detailed calculation for each process can be found in Table S2. 1. In general, the mass balance principle is applied to determine the resource efficiency, outflow, inflow and losses of each unit process, and the relationship of those four parameters is shown in Fig. S2.1. Given two parameters, the other two parameters for a unit process can be obtained based on the principle of mass balance. Notably, for the application of the mass balance approach, some input parameters should be given exogenously through four approaches (i.e. statistical data, technology analysis, or mass flow allocation), which are marked in Table S2.1. Herein, the technical analysis refers to a literature investigation on the technical features (e.g. yield rate, energy use, energy sources) of the studied technology based on the related technical reports, patent documents, or publications at that time.  Note: the world steel yearbook data 28 represents all yearbook from 1978 to 2017 in the official website of world steel association.

S14
As listed in Table S2.1, the data sources for mining RE and mass allocation in rolling and finishing stage should be further clarified as follows: (a) Mining RE. There is no statistical data regarding the RE of mining. Herein, its historical trend is estimated by the following steps: Firstly, as shown in Fig. S2.2A, this study collected the RE data for different countries in 2000 from 4 , and its ore grade data from USGS mineral yearbook. It is found that there is a high correlation between the ore grade and RE. The second step is to obtain the historical ore grade trend, which is also a lack of statistical data. Hence, based on the investigation from 41 for the period from 1905 to 1925 and study 42 for the period from 2000 to 2015. The estimated trend of ore grade is shown in Fig. S2.2C.
Finally, as shown in Fig. S2.2D, the historical trend of RE can be obtained based on the ore grade trend and its relationship with RE.

S15
(b) Steel product mass allocation. This study quantify the steel flows from casting to end-use sectors based on the studies from 2 , the relationship of each flow is mapped in Fig.S2.3A (the red figure represents the material flow as noted in Fig. S2.3A). This study assumed all the cold rolling and finishing stage as one process to produce final cold rolling (CR) products, and the production flow from ingot casting (i.

S16
(2) Fabrication, In-use and End-of-life stage Those three stages are treated herein as they are more application-specific rather than material-specific as in the material supply system. In link with other steel MFA studies 3 , the end-use applications are mainly divided into four sectors: Construction, Vehicles, Machinery, and Daily goods. The corresponding parameters, like market share, lifetime, and resource efficiency, are specific to each other.
(a) Fabrication. The product fabrication stage transfers the raw materials from either primary or secondary production into different products for use. For the material in sector n, the inflow is determined by the allocation model from 2 . Meanwhile, the fabrication rate (FR) are assumed to follow the current level from 2 , which is shown in Table S2.2. The scrap from fabrication is named promote scrap, which is assumed to be fully recycled and treated in the secondary production stage. (b) In-use Stage. The products from fabrication stage finally enter its corresponding in-use stage to provide services as in-use stock. The inflow into the in-use stock equals the final products from fabrication stage. As for the outflow, it follows the lifetime distribution scheme 3 : where, ( , , ) is the probability density of the lifetime distribution function; t is the quantification time step; is the lifetime of this product sector; is the standard deviation of lifetime; is the end of the studied period; is the starting time. The ( ) is the initial in-use stock in the year 1900 based on the study from 44 . The lifetime distribution could be Normal, Delta or Weibull distribution, and the normal distribution and its lifetime is adopted from 3,29 in this study.
The specific parameters for fabrication rate are obtained from 3,29 . Meanwhile, 10% of outflow in the construction sector is assumed to be accumulated as obsolete stock, which refers to the products which are out of service and not accessible for recycling 3,29,45 . The basic settings for those parameters are shown in Table S3.
(c) Scrap, Recycling and Losses. Three types of scrap (i.e. home, new and old scrap) are generated from the anthropogenic iron cycle. As shown in Fig. S2.4, this study follows the conventional definitions of those scraps 46 : home scrap refers to scrap generated from casting, or foundry production, which can be internally recycled inside the material product sites. New scrap (or prompt scrap) refers to the scrap generated from during the fabrication of metal products, which is transferred to the scrap market for further recycling. Old scrap (or EoL scrap, postconsumer scrap) refers to the scrap generated from the products, which enters to the end of their service life.

S17
The recycling stage includes a series of processes, e.g. collection, sorting, and separation. According to the 46 , the end-of-life recycling rate (EoL-RR) equals to the recycled Old-scrap(c2 in Figure S2.4) divided by the old scrap generation. Herein, the old scrap generation equals to the outflow in the equation S1, and the new scrap recycling rate is assumed to be same with EoL RR. Based on mass balance, the recycled new and old scrap can be obtained. Furthermore, this study also quantifies the annual resource loss from each life cycle stage, as shown in Fig. S2 A bottom-up approach was chosen for quantifying emissions of greenhouse gases (GHG) from steel production. The benefit of the bottom-up approach is that it allows for disaggregating steel production into different steel production processes. (e.g., mining, blast furnace, cold rolling, etc.). This allows for modelling and expressing the technological development of the different processes between 1900 and 2015. Such specific differentiation cannot be obtained using top-down approaches, such as environmentally extended input-output databases.
The bottom-up approach was modelled using the life-cycle assessment (LCA) software SimaPro version 9.0.0.35. General activities in the steel processing sector were modelled using information from the Ecoinvent life-cycle inventory database 47 coupled with specific process information as described in Section S2.1 to obtain a comprehensive overview of the processes involved in steel processing between 1900 and 2015.
The coupling with information from Section S2.1 is particularly important for processes, which are not used anymore and, not covered in the ecoinvent database that primarily covers processes that are currently part of the steel processing industry. Because the information given in Section S2.1 is generally for specific inputs and outputs of the process (e.g. electricity or coke use), process specific details, such as direct carbon emissions associated with running the process, were not available (e.g. CO2 emissions as a result of carbon removal from the iron ore and generation of residual waste for treatment). For this, we used the available process information from ecoinvent 47 and extrapolated this to processes with a similar function in the steel processing value chain, in order to construct more complete unit processes for all activities involved in steel processing which include both direct and indirect emissions and resource uses.
A full presentation of the life-cycle inventory used for modelling the 19 steel processing activities is given in Supplementary Data 1. This includes all direct emissions and resources uses for each process as well as all process inputs required for the functioning of the steel processing process, such as electricity generation, coke production, oxygen production, etc.
Herein, several essential information regarding this historical calculation is introduced as follows:

S2.2.1 Historical trajectory of energy efficiency
The statistical data related to the annual energy use of each studied production processes are not available.
This study estimated the annual energy inventory of each technology based on the typical value and their trajectories in the history, which are obtained in two steps: firstly, this study collected the trend of energy use from various published studies and presented in the Fig. S2.5, where the key reference is summarized in   In this section, the system boundary for of each studied technology is illustrated in Fig.S2.6-S2.15. The system boundaries indicate the main processes as well as inputs and outputs from auxiliary processes. Hereby, Fig.S2.6-S2.15 provide an overview of the processes covered and used for quantifying the technology specific GHG emissions for each studied technology. The system boundaries are divided into a foreground and a background system. The foreground system includes technology specific processes for which specific modelling choices have been made about the process, such as historical development of the technology. This is presented in Section S2.1 and S2.2. The background system includes more generic unit processes from ecoinvent 3.1 that are used as inputs for processes in the foreground system. Please see Supporting Material 2 for a full overview of the life-cycle inventory and processes used for modelling each of the 19 studied processes covered in this study. (Upstream processes in the steel life-cycle are indicated with dotted line borders. GHG emissions pertaining to such upstream processes is taken into account in the system boundary for that particular process and, thus, not included in the GHG emission quantification for subsequent downstream processes.)

Foreground system
Background system  to such upstream processes is taken into account in the system boundary for that particular process and, thus, not included in the GHG emission quantification for subsequent downstream processes.)

Foreground system Background system
Blast furnace pig iron production Pig iron  Fig. S2.10 System boundary for GHG emission quantification of iron foundry process. Upstream processes in the steel life-cycle are indicated with dotted line borders. GHG emissions pertaining to such upstream processes is taken into account in the system boundary for that particular process and, thus, not included in the GHG emission quantification for subsequent downstream processes. GHG emissions pertaining to such upstream processes is taken into account in the system boundary for that particular process and, thus, not included in the GHG emission quantification for subsequent downstream processes.)

Foreground system
Background system  to such upstream processes is taken into account in the system boundary for that particular process and, thus, not included in the GHG emission quantification for subsequent downstream processes.)   GHG emissions pertaining to such upstream processes is taken into account in the system boundary for that particular process and, thus, not included in the GHG emission quantification for subsequent downstream processes.)

Foreground system
Background system

.3 GHG emission calculation
Based on the technology and time differentiated inventory of GHG emissions, the total GHG intensity for each steel production technology was estimated according to Eq. S3.
Where E i is the GHG emission intensity of steel production technology i [kg CO2-eq / kg output] at year t (see Figure S2. 16). E x,i is kg emission of GHG x per kg output from steel production technology i at time t. GWP100 x is the global warming potential [kg CO2-eq / kg GHGx emitted] for GHG x 66 (see Table S2.4).
The development in total GHG intensity [kg CO2-eq / process output] over time for each steel processing process is shown in Figure S2.6.
The total GHG emission per steel production technology per year was estimated as: Where mGHG i (t) is the total emission of CO2-eq in year t from steel production technology i. m i (t) is the total output from steel production technology i at time t estimated using the dynamic MFA model. The sum of all CO2-eq emissions from all steel production processes in year t gave the total emission of CO2-eq from steel production in year t:  The uncertainty assessment will be introduced in Section S2.3. GHG emissions from steel processing processes generally decrease over time as the technology is improved, as is reflected in total GHG intensity for steel manufacturing. However, increases in GHG emissions were observed for mining, direct iron reduction, and electric arc furnace (EAF). The increase for mining is due to the increasing need for energy to extract the iron ore from the mines. Changes in GHG intensity for other processes is either due to change in the performance of the technology, e.g., via increase energy or material efficiency, or because the energy grid mix needed for the process changes over time. For instance, a relatively large share of the global electricity mix came from hydropower (22%) in 1980 67 . In 2000 to 2015, the share of electricity from hydropower was only about 16%, and a proportionally larger share of electricity came from fossil fuels 68 . For this reason, several processes also show an increase in GHG emission intensity from around 2009. This increase is due to an increased share of electricity and heat coming from coal and natural gas. The historical development in electricity and heat generation is shown in Table S2.5 and Table S2.6, respectively.

S2.3 Uncertainty analysis
The data sources and uncertainty levels of those exogenously given input parameters are listed in Table S2.7 for material flow quantification and Table 2.8 for greenhouse gas emission quantification.

S3.1 Material Flow Results
In this section, the related activity data sets for emission calculation are presented in the first stage.
Afterwards, other relevant data series (i.e. in-use stock trend, end-of-life scrap trend, and EoL (End-of-life) recycling rate, etc.) for our analysis are presented.
(1) Production activity data sets The impact analysis requires the outflow data of 19 studied processes, which are obtained either from official statistical data sources or from our material flow model.
As shown in Fig.S3.1, there are 9 parameters obtained directly from official statistical data sources, including mining production from 13,30 , DRI production from 19 , EAF, OHF, and BOF production from 23,28,34 , and Puddling, Bessemer, and Crucible from 23,34 . Given those that data are obtained from the official statistical data sources, the uncertainties of those trends are very low.  impact on the total results is quite low, given their relatively small magnitude and low impact intensity.
Consequently, the results from material flow analysis are acceptable for impact analysis, and a Monte-Carlo analysis will be conducted for the uncertainty analysis.

S44
(2) Data sets for in-use stock, scrap, and recycling rates This study estimated the annual amount of steel flow and stocks in the fabrication, in-use and end-of-life stage. The major results regarding in-use stock, scrap amount and recycling rate is shown in Fig. S3.3.

S3.2 GHG Emission Results
The historical GHG emisison associated with global steel production is presented in Fig.S3.4, while the detailed tracjectory for each technology is shown in Fig.3.5 and Fig. S3.6.

S3.3 Results evaluation
This sub-section compares our results with other studies. Notably, most of the material flow results (in Fig. S3.1-3.2) in the material production stage is directly obtained from official statistics or mass balance. Hence, the accuracy of those results can be guaranteed. Herein, the Pearson correlation is adapted for comparing the production activities data in Fig. S3.7 and this study compared the results in other life cycle stages (i.e. in-use stock, scrap, etc.) with other studies, as shown in Table S3.1. Furthermore, the greenhouse gas emission of our study is compared with other published ones in Fig. S3.8.

S3.4 GHG intensity results
The GHG intensity of each production routes are presented in Fig. S3.9. We firstly applied the LMDI decomposition method to clarify the contribution of volume and intensity on the absolute emission change, and the method is described in the following equations.
The total emission ( ) is equal to the volume ( ) times intensity ( ), and then the change of total emission from t-∆ to t can be decomposed into these two factors as follows: Where the absolute change of those three factors are: ) (S10) To check whether the efficiency (or the intensity) improvement can lead to the absolute reduction of total emission, we further applied the relative index decomposition to trace the interaction of those factors (i.e. emission, volume, intensity, and the negligible interaction of volume and intensity) as follows: Where each term is the relative emission change (in percent) along the designed periods and such decomposition analysis can eliminate the scale factor and to reflect the contribution of each factor more straight compared to the pervious LMDI approach.
For a relatively substantial investigation, we applied the above decomposition analysis for every five years, when the figures are in a cumulative format in each period. The detailed results are shown in Table S4.1. We further investigated the correlation relationship for those three factors, and the results are shown in Fig.S4.3.   We propose six scenario sets to indicate the future pathways of required technical and material efficiency improvement to follow the 1.5DS pathway, the details of which are described in Table   S4.2 and Fig. S4.6. As for future demand trends in Fig.S4.6a, IEA 85 has published their estimates of steel production from 2011 to 2050 with two scenarios -stated policy scenario and sustainable development scenario. The stated policy scenario describe future steel demand under the business as usual. With the implementation of material efficiency strategies, the future steel demand can be reduced while maintaining the same service, and the projection is presented in sustainable development scenario.
As for the carbon budget estimation under IEA 1.5DS, we obtained this level based on the S59 efficiency trend from IEA 1.5DS (i.e. decreasing from 2.38 t CO2-eq/t in 2010 to 0.32 t CO2-eq/t in 2050 in Fig. 4.7b). Under this efficiency condition, the highest demand scenario (i.e. stated policy scenario) in Fig.S4.7a is chosen to obtain the high variance of such a carbon budget, totaling 106 Gt CO2-eq from 2010 to 2050. One recent study 86 has reviewed the existing framework and amounts of the carbon budget for all human activities after 2018 to hold warming to 1.5 °C above pre-industrial levels (50% chance), which stays 420-580 Gt CO2-eq. Given the steel sector only accounts for 7-9% of global GHG emission, the carbon budget of 106 Gt CO2-eq should be halved according to this allocation (18%-28%), indicating a more stringent carbon constraint on future steel production. The GHG emission pathways under our proposed scenarios are shown in Fig.   S4.7c. This subsection summarized the national policies in decarbonizing steel and other energy-intensive industries from the IEA policy and measure databases in Table   S4.3, and most international emissions reduction policies were linked to efficiency, energy use, and carbon within production sites. There is little apparent attention from those policies focusing on the role of material life cycle on the carbon reduction in production stage.   Using hydrogen (instead of coal) for the direct reduction of iron oxide/ore (H-DR), combined with an electric arc furnace (EAF).
Reducing agent used-the main source of hydrogen is the electrolysis of water to produce hydrogen.
The electricity used in electrolysis of water comes from clean energy power stations such as water power and wind power. The project will produce 0.5 Mt/a DRI using hydrogen, which is extracted from the co-products of a natural gasbased process to make vinyl acetate. The technology is based on the usage of hydrogen plasma. Thereby, hydrogen is used as the reduction agent for the iron ore while its plasma state offers the thermal energy for melting the metallurgical iron. The utilisation of hydrogen as the reduction agent inheres the advantage that only gaseous water remains as by-product. The Nucor plant will use energy produced by Evergy, including from a new wind farm, to power electric arc furnaces that will melt scrapped steel and turn it into new, recycled steels which set to be the first U.S. steel plant to run on wind energy. Reducing yield losses in manufacturing (e.g. sheet metal in the automotive industry) would reduce material demand and in turn emissions from material production. Additive manufacturing, by its nature leads to minimal material losses compared to processes that cut an object from larger pieces of material.
There are plans to test the demonstrator in a real test as the project progresses.

S72
S4.7 Summary of material efficiency strategies related to steel use Material efficiency 125 refers to make more out of less material use, and the detailed measures have been widely examined for steel (around six types of strategies, i.e., Less Material Same Service, More Intensive Use, Lifespan Extension, Fabrication Scrap Diversion, Reuse of End-of-Life Scrap, and Yield Improvement) 48,[125][126][127] that are considered as essential decarbonization strategy according to various assessments 128 . Here, we collected the detailed implementation strategies and some representative cases for the main end-use of steel in Table S4.5. Assemble different components precisely instead of high-labor-cost welding, 3D printing, etc.
Aggressive light weighting of beverage can result in a 35% reduction in material requirement, etc.

More Intensive Use Principle
 Make the building publicly accessible and improve the housing leasing system to reduce the demand for housing.  Reduce the demand for private transportation through reasonable scheduling of public transportation and the availability of transportation.  By renting production equipment instead of self-purchasing equipment, the idle rate of machines can be reduced. At the same time, enterprises can avoid the risk of technological backwardness and improve the efficiency and safety of construction.  Using public goods to replace personal low-frequency goods can increase product utilization and reduce the mass production of the product. Increase the utilization rate of trains, Car-sharing, Ride-sharing, etc.

S73
Construction machinery rental service, etc.
Shared power bank. Shared bicycle, etc.

Lifespan Extension
Principle  Making components more durable, cascading products between users with different requirements or upgrading products to extend the useful life of their embedded materials  Regularly check the operating conditions of vehicles and mechanical equipment to extend the service life.  Use personal consumer products in a suitable environment to keep them in good condition.

Cases
Clear the roof after snowstorms, etc.
New metal is added to the rail in-situ in thin layers, and the rail can be restored with high strength, etc.
Regularly clean up mechanical equipment to extend the service life of mechanical equipment, etc.
Use personal consumer products in a suitable environment, avoid bumps, and avoid using them in rustprone environments, etc.

Fabrication Scrap Diversion
Principle Building materials will be surplus due to overordering. These extra materials can be made into other equipment.
The unused part of the vehicle manufacturing process can be collected and processed into other equipment.
Steel waste that does not become a product in the production process (such as trimming, trimming, etc.) Consumer products are massproduced, which require preordered materials. The order quantity of raw materials is usually more than the actual quantity required.

Cases
Extra pre-ordered steel bars can be made into a pedal net, etc.
Abbey Steel in Kettering purchased blanking skeletons and other trim (such as the window cut-outs in door panels) from car manufacturers, then cut regular shapes, supply them as blanks to firms making small parts, etc.
The roof trusses of the main stadium in 2012 London Olympics: are made from over-ordered oil and gas pipeline, etc.
The leftover material from making white goods can be used to make tables and chairs, etc.
Principle  Making components more durable, cascading products between users with different requirements or upgrading products to extend the useful life of their embedded materials  Regularly check the operating conditions of vehicles and mechanical equipment to extend the service life.  Use personal consumer products in a suitable environment to keep them in good condition.

Cases
Steel-framed buildings may be changed due to planning policies, and steel from Shipbreakers will dismantle scrapped ships and obtain large amounts of steel and other metals Recycling of mechanical processing equipment.
Reprocessing and use of consumer goods S74 Strategies

Construction
Transportation Machinery Consumer goods which can be used again. The British Construction Steelwork Association's new headquarters building was constructed from used steel

Reuse of End-of-Life Scrap
Principle Building materials will be surplus due to overordering. These extra materials can be made into other equipment.
The unused part of the vehicle manufacturing process can be collected and processed into other equipment.
Steel waste that does not become a product in the production process (such as trimming, trimming, etc.) Consumer products are massproduced, which require preordered materials. The order quantity of raw materials is usually more than the actual quantity required.

Cases
Extra pre-ordered steel bars can be made into a pedal net, etc.
Abbey Steel in Kettering purchased blanking skeletons and other trim (such as the window cut-outs in door panels) from car manufacturers, then cut regular shapes, supply them as blanks to firms making small parts, etc.
The roof trusses of the main stadium in 2012 London Olympics: are made from over-ordered oil and gas pipeline, etc.
The leftover material from making white goods can be used to make tables and chairs, etc.

Yield Improvement
Principle  Reasonable planning can reduce product waste during construction.  Using high-grade steel can improve performance while avoiding the use of large amounts of low-quality steel.  Manage production materials well to avoid corrosion in their storage.

Cases
Qube Design rationalizes the layout of reinforcing steel.
High-strength steel for automobiles can subdivide component size and structural weight, save steel consumption, and is higher than ordinary-strength steel in terms of overall stability.
A novel technology for rolling create beams using one-third less metal than standard I-beams, Lean manufacturing, etc.
Optimize supply chain control links and maintain a reasonable inventory