- Email: [email protected]
Benchmarking aluminium die casting operations
Resources, Conservation and Recycling 52 (2008) 1185–1189
Contents lists available at ScienceDirect
Resources, Conservation and Recycling journal homepage: www.elsevier.com/locate/resconrec
Benchmarking aluminium die casting operations A. Tharumarajah ∗ CSIRO Sustainable Ecosystems, 37 Graham Road, Highett, Victoria 3190, Australia
a r t i c l e
i n f o
Article history: Received 4 March 2008 Received in revised form 10 June 2008 Accepted 17 June 2008 Available online 31 July 2008 Keywords: Die casting Aluminium Benchmarking Recycling Environment
a b s t r a c t Increasing demand in world automotive markets for aluminium die cast components is creating significant opportunities and challenges for the Australian industry, which is positioning itself as a global player. To meet these challenges, the industry is continuously seeking to improve its overall resource efficiency that can result in the reduction of cost and impact on green house gas (GHG) emissions. In order to understand and evaluate the current position, this study benchmarks the use of aluminium and high-use operating resources of a large representative aluminium high-pressure die casting (HPDC) facility. By modelling the complex web of product, recycling and waste flows, resource efficiencies, costs and GHG impacts of the considered resources are computed. The central focus of this study is the in-house recycling flows of aluminium, normally considered good practice, though it can have negative implications on resource efficiency, costs and GHG impact. In fact, as revealed by this study, the recycling losses contribute as much as around 49% of the total aluminium melted adding about 44% to the cost of manufacture and 50% of the GHG added in production. Using the insights obtained, the technological and other systemic factors that contribute to the losses are identified and areas of improvements are suggested. Crown Copyright © 2008 Published by Elsevier B.V. All rights reserved.
1. Introduction The attractiveness of die casting is its ability to make near net-shape parts with tight tolerances and requiring little or no machining. The more applied of the different die casting processes is high-pressure die casting (HPDC) with high rates of production. The automotive industry uses an extensive range of aluminium HPDC parts including transmission housings, cylinder heads, inlet manifolds, engine sumps as well as decorative trim. This trend is increasing as replacement of steel parts with lighter aluminium HPDC parts grows (Young and Eisen, 2000). To meet the challenges of competition, the industry is continuously seeking to improve its overall performance. While costs, delivery and quality have been the traditional determinants of this, growing concerns about environmental burdens (green house gas (GHG) emissions) created by manufacturing has the industry seeking to reduce the GHG emissions as well. Notably, HPDC processes can be energy and water intensive with high percentage of losses of high cost aluminium as trim wastes. In order to benchmark the flows and wastes, this study considers the metabolism of physical flows in a representative die casting industry that has an estimated 10% of the Australian market. The study focuses on top five of the resources used, namely aluminium
and associated operating resources, viz. water, electricity, gas and die-lubricant (die-lube). Using input–output mapping, the product, waste and circulating (recycling) flows are mapped and the amounts are estimated using material flow analysis (MFA, Brunner and Rechberger, 2004). The estimation is accompanied by calculation of direct and added costs, and the GHG impacts. The central focus being the in-house recycling flows of aluminium. Normally, such recycling is considered good practice, though it can have negative implications on resource efficiency, costs and GHG impact. This estimation reveals the extent of the resource efficiencies that are further examined to investigate the influencing factors. These factors, normally restricted to the die casting process, are expanded to include social factors such as training, maintenance, quality assurance, and shop-floor information and management systems. In Section 2, analysis of the physical flows and their estimation is given. This is followed by the analysis of material efficiencies, costs and environmental impacts from these flows. Section 4 discusses the results of benchmarking and examines technological and other factors that influence the losses and suggests avenues for improvement. Finally, conclusions are drawn with some pointers to future work. 2. Resource flows in HPDC
∗ Tel.: +61 3 9252 6458; fax: +61 3 9252 6249. E-mail address: [email protected].
Investigating usage and cost efficiencies (or intensities) of aluminium (alloy ADC12) and associated operating inputs requires a
0921-3449/$ – see front matter. Crown Copyright © 2008 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.resconrec.2008.06.007
1186
A. Tharumarajah / Resources, Conservation and Recycling 52 (2008) 1185–1189
study of the flows of these through the various operations in the die casting plant. Operating resources, in particular, can be the numerous lubricants (such as die-lubes, tip lubes), chemicals (anti-rust, detergents, silver grease), water, fuels (for maintenance), natural gas, electricity and so on. In this study, the flow of resources is limited to aluminium and operating inputs of electricity, natural gas, die-lube and water used in the production. Usage of these operating resources is largely dictated by the parts produced. To understand the flows, Fig. 1 provides a schematic of the flow of aluminium through the plant and the dependent flow of the operating resources. The flow of aluminium through the plant can be described by the following distinct stages: • Aluminium for production comes from three sources: as hot metal deliveries (HMD), recycled process scrap (in-house, from rejects and process losses) and ADC12 ingots that are input to recycled aluminium. • Melting and holding furnaces store the molten HMD and recycled aluminium. • Parts are cast at the HPDC machines of different tonnages (800 T × 5, 1250 T × 4, 2250 T × 2 and 2500 T × 2) depending on their design with some capability overlap between the machines. • Excess aluminium in parts is trimmed and the part is finished (some parts undergo machining) and the parts, after testing and inspection, are shipped to the customers. The in- and out-flows of aluminium and investigated operating inputs, shown in Fig. 1 can be complex. In particular, a number of waste streams (i.e. non-productive or not saleable product outputs) of aluminium and operating inputs occur right throughout the process. Significant of these (for aluminium) include: • Oxidation losses at melting and holding furnaces (sold to aluminium recyclers). • Process losses (such as warm-ups, mis-runs and others) at the casting process (mostly recycled).
Table 1 HPDC Al flows by weight for the data period Category
Weight (tonnes)
Aluminium processed Aluminium waste sold Aluminium waste recycled in-house Aluminium parts shipped
15,611 507 7,663 7,441
• Yield losses (such as runners, biscuits and flashings, mainly due to die design) that is trimmed (recycled). • Turnings from machining (sold to aluminium recyclers). Wastes from operating inputs include waste water (mixed with die-lube waste) that is treated and disposed as trade waste. Energy losses from electricity and gas use are not captured. Using material flow analysis the inputs and outputs at the various production stages and at supporting processes are mapped. However, the depth of detailed mapping differs due to data availability and the purpose for which analysis is undertaken. For instance, flows of aluminium is tracked at the part level (due to part-dependent differences in shot weight, yield and process losses, and general availability of data), whereas, electricity is categorised by consumption by one or the other processes or equipment. 3. Usage and cost efficiencies 3.1. Aluminium (HPDC) 3.1.1. Usage efficiency The flow (by weight) of aluminium used in HPDC is computed from design data for the individual parts and daily production data from April 2006 through March 2007 for the machine groups. This data is used in computing the material efficiency. Normal efficiencies consider the material in a product and associated waste (that leaves the plant). This efficiency, called here as 1st pass material efficiency, is calculated from the flow data (see
Fig. 1. Flow of aluminium and others through HPDC process.
A. Tharumarajah / Resources, Conservation and Recycling 52 (2008) 1185–1189 Table 2 Cost breakdown of aluminium losses
Table 1) as follows Ist pass material efficiency =
Al-Sh × 100% = 93.87% Al-Sh + Al-Ws
where, Al-Sh is the aluminium parts shipped and Al-Ws is the aluminium waste sold (furnace losses, oily scrap and machining losses). This efficiency, calculated using data in Table 1, indicates that a very low percentage of ∼6.1% of the product material ends up as waste. But, this is not the case since the HPDC process produces a significant amount of hidden aluminium losses, i.e. in-house recycled aluminium that circulates. These losses, nevertheless can contribute substantially to the cost of making the products. Thus, a better measure is to consider the material efficiency from the inputs of aluminium including that which is recycled in-house from process losses. This operational material efficiency is calculated as operational material efficiency =
1187
Al-Mt − Al-Ws − Al-Wr × 100% = 47.7% Al-Mt
where, Al-Mt is the aluminium processed (i.e. total melt), Al-Ws is aluminium waste sold (furnace losses, oily scrap and machining losses), Al-Wr is aluminium waste recycled in-house (yield and process losses) and Al-Sh is the aluminium parts shipped. Comparing the material efficiencies with and without in-house recycling reveals that while only 6.1% of the product leaves the plant as waste (1st pass efficiency), a substantial percentage of 52.3% of the aluminium processed ends up as losses when operational material efficiency is considered. Comparing more detailed operational material efficiencies among machine groups, the highest is for 2500 T (∼52%) and the lowest is for 800 T (∼42%). The lower values for 800 T are mainly due to relatively lower process losses (warm-ups, etc.), though the yield losses (influenced by the part and die design) are higher. Furnace loss is assumed constant at 2.5%. 3.1.2. Cost efficiency For the purpose of computing cost efficiency, waste includes all resources which are non-productive or not saleable as a product (Sustainability Victoria, 2008). That is, it includes all the aluminium losses that occur whether they are recycled in-house or sold to outside recyclers. There are two types of costs associated with aluminium waste (both recycled and sold): • Added cost of waste: production costs incurred in processing aluminium through the plant before waste occurs. For example, cost of melting for oxidation losses or cost of casting before warm-up loses. This cost is estimated for each stage of production from the total cost of manufacture, i.e. cost of materials (other than aluminium), direct labour, other variable (includes other materials, consumables, etc.), indirect labour, depreciation and others. • Cost of handling and transport of waste: these costs are associated with waste sold and are estimated from actual costs.
Loss category
% of total cost of losses
Process losses Yield losses Waste sold Cost of additional Al assigned to waste Total cost of losses
29.8 66.9 0.6 3.3 100.0
3.2. Usage intensities of operating resources The usage intensity of operating resources, viz. electricity, gas, die-lube and water is estimated by category of use. Intensity measure used is consumption per tonne (or kg) of aluminium product sold. In order to allocate their usage for production activities, the category of uses are further analysed and an estimate is made for apportioning to production. Consider the case of electricity, Fig. 2 shows the major categories of its use by consumption and percentage. The breakdown of the intensity of electricity consumption reveals, approximately 40% is consumed in direct production (HPDC machines, CNC machining centres and heat treatment) and a higher percentage of about 47% on production services (air compression, cooling system and 95% lighting). That is, a total of 87% of electricity consumption is in production. Table 3 indicates the usage of all the resources considered in this study. It shows percentage allocated to production, intensity of use per tonne of aluminium sold and % cost added to production. Reviewing the intensities, the intensities for electricity and water are quite high, though without a comparative value it is difficult to judge efficiency of usage. Nevertheless, use of around 4.3 l of (drinking) water per kg of aluminium sold is of concern. Taking the analysis a step further, respectively, around 0.6 l and 1.6 l of water
1 − total cost of losses × 100 = 55.9% total manufacturing costs
A breakdown of the cost of waste is given in Table 2. Cost percentage attributed to the losses takes into account the production stage at which the losses occur. For instance, processes losses (i.e.
13.1 29.5 0.25 1.46 44.1
losses due to warm-up and mis-runs, and other rejects during casting) occur at the casting stage, and hence the added cost of these losses includes cost of melting, holding and casting, and transport and handling. Similarly, yield losses are at the trimming stage. As for waste that is sold, the losses include furnace oxidation losses, oily scrap at HPDC and machining losses. Waste sold generates revenue, but the net result is a cost to plant. The highest source of added costs of waste is yield losses, which largely depends on the part and die design, and thus can be difficult to reduce in the short-term. On the other hand, there is potential to reduce process losses that amount to approximately 30% of the cost of all losses or about 13% of the manufacturing cost. The cost of additional aluminium to compensate for the shortfall (aluminium parts and waste sold are greater than the aluminium recycled inhouse) is small at 1.46% of manufacturing cost.
Resource cost efficiency of aluminium is computed using estimated cost of losses (shown as a % of manufacturing cost in Table 2) as follows resource cost efficiency =
% of total cost of manufacture
Fig. 2. Electricity consumption by category of use.
1188
A. Tharumarajah / Resources, Conservation and Recycling 52 (2008) 1185–1189
Table 3 Usage and cost of operating resources
and yield losses. Thus, a fair proportion of around 49% of the total GHG is added for the losses.
Operating resource
% allocated to production
Usage intensity/ tonne of Al sold
% of manufacturing cost
Electricity (kWh) Natural gas (MJ) Die-lube (L) Water (L)
87.5 100.0 100.0 78.4
1.54 0.02 0.01 4.32
1.33 1.02 0.65 0.11
is expended on process and yield losses (process and yield losses account for around 13% and 36% of total aluminium processed in the plant, respectively). Interesting aspect of the cost operating resources is that they only contribute a small percentage of around 3% to the cost of manufacture with the highest contributor being electricity. 3.3. Environmental impact In computing the environmental impact, i.e. GHG, only contributions by way of additional aluminium bought from outside and usage of operating resources is taken into account. The additional aluminium compensates for the balance between what is bought, recycled, sold as product and sold as waste. Thus, it is considered as part of the waste. Total amount of additional aluminium is around is 224 T or 0.03 kg per kg of aluminium sold by the plant. In terms of GHG impact this translates to around 0.5 kg of CO2 equiv. per kg of aluminium sold, assuming a GHG impact of 16.5 kg CO2 -equiv. per kg (this assumes a composition of 50:50 primary to secondary aluminium produced in Australia, Ramakrishnan et al., 2003). The second source of added GHG is the operating resources considered in this study. Based on the usage intensities of these resources (see Table 3) GHG impact in terms of kg CO2 -equiv. for each of the operating resources is calculated. As shown in Table 4, of the ∼4.1 kg CO2 -equiv. of GHG added in producing 1 kg of finished casting, around 88% is contributed by the operating resources, with electricity and water contributing around 51% and 37%, respectively. The higher impact of electricity is due to the use of brown coal that has a very high global warming potential for generating electricity. Aside, the cost of these two resources is a fraction of total manufacturing cost (Table 3), and hence incentive to reduce these comes from their environmental impact, rather than cost. In the case of water, water used in production does not have a high GHG burden. The burden is added in treatment with impacts coming from use of caustic soda, etc. (Tharumarajah et al., 2003). Contribution from natural gas is negligible due to its lower impact and less usage intensity and so is die-lube. The table also shows the portion of the added GHG that can be attributed to process and yield losses that are recycled. Assuming a linear proportion, the impact due to these can be calculated, i.e., respectively, 13% and 36% of the total GHG are attributed to process Table 4 Greenhouse gas impact of operating resources Item
GHG (kg CO2 -equiv.) Per kg of Al sold
Allocated to recycled process losses per kg Al sold
Allocated to recycled yield losses per kg sold
Additional Al Electricity Natural gas Die-lube Water
0.50 2.15 0.00 0.01 1.54
0.07 0.29 0.00 0.00 0.21
0.18 0.78 0.00 0.00 0.56
Total
4.19
0.56
1.52
4. Discussion on improvement The estimation of aluminium and operating resources reveals the extent to which aluminium losses contribute to costs and GHG impacts. While most of these losses are recycled, nevertheless, the hidden cost of these losses can be high (see Table 2) and lessening of these losses can reduce the waste (sold). Additionally, the highusage operating resources (electricity, gas, water and die-lube) considered here have very high dependence on the aluminium processed. Thus, savings in aluminium losses is bound to save on these too. In improving resource efficiency and reducing cost of losses (see Tables 1 and 2), two principal avenues can be considered. Yield losses occur mainly due to excess metal in runners and flashings. These losses depend on the part and die design. However, recent technological advances based on novel runner design and semisolid feed system, such as advanced thixotropic metallurgy (ATM, Gunesegaram et al., 2007), can reduce the shot weight and improve porosity of cast parts. Application and acceptance of such technologies require extensive industry trials before becoming main stream technologies and hence is considered to be a medium–long term opportunity. On the other hand, process losses are a result of casting machines being unable to maintain operating conditions during production, such as warm-up and mis-runs. It can also be due to machine maintenance. Improvements in this area offer a viable short-term option. Discussion on improvements is thus focussed on factors that underpin these losses. The two most causes of process losses are warm-ups and misruns. Warm-up losses occur due to the need to warm-up the die by using the heat from molten aluminium during start up phase of producing a part. Warm-up castings are rejects and are recycled. The number of warm-up castings before acceptable parts are produced depends on many factors including the condition of the machine to maintain an optimum temperature and pressure (to inject the molten metal in to the die cavity). It can also be affected by the concentration and amount of warm-up release agent used when the die is cold. A mis-run or cold shut, on the other hand, occurs when a casting lacks completeness due to the failure of the metal to fill the die cavity. This may be due to poor fluidity of the metal, machine unable to maintain temperature and pressure optimum for casting or due to part and die features such as too thin sections and wall thickness and improper gating system. Thus, the primary reason of these losses can be attributed to the condition of the casting process and/or the part and die design. To add to these issues, the nature of the casting process is such that when disruptions to product cycle are experienced, the optimum conditions tend to deteriorate. These instances include breakdown, die-changes, process cleanup and others. Thus, there may be other secondary issues other than those with machines, dies and quality of aluminium. In fact, the secondary issues can be much deeper, such as inadequate operator training, lack of timely information and delays in reacting to problems, and lack of knowledge of processes and idiosyncrasies of parts, in particular new parts. Thus, as shown in Fig. 3, they become influential in trying to improve product quality, though more often than not improvements are directed at dealing with problems at the machines, dies or melt quality. The danger of such myopic approaches is that the secondary issues can fester and tend to become systemic affecting overall productivity of the shop-floor.
A. Tharumarajah / Resources, Conservation and Recycling 52 (2008) 1185–1189
1189
losses can have a positive flow-on effect in reducing the consumption of these too. Additional benefits may arise in the reduction of greenhouse gases, in particular through the reduction of electricity consumption. In essence, a study to benchmark flows can reveal many insights when circulating in-plant recycling flows are considered. In this case, in-house aluminium recycling though considered a benefit to the operation, nevertheless adds substantially to the cost and GHG emissions. In fact, quantification of these impacts can be an eye opener in exposing the inefficiencies which is otherwise hidden, in particular GHG contributions of low cost resources. The study also briefly examined process and softer systemic factors for reducing the losses. Some of the latter, such as operator knowledge can be critical, in particular when quality is affected by the complexity and idiosyncrasies of the parts produced. Others, such as shop-floor management and information systems also may act as impediments when quick diagnosis and action are required. Fig. 3. Primary and secondary factors influencing part quality and shop-floor efficiency.
Acknowledgements
Other than savings in aluminium, opportunities exist for separation, recovery and reuse of die-lube from waste water. The analysis has revealed the savings in pursuing this avenue. Such recycling along with rain water harvesting could lead to self-sufficiency in water.
The author wishes to thank Sustainability Victoria and Cooperative Research Centre for Casting and Solidification Technologies for their support in conducting this study.
5. Conclusion
Brunner PH, Rechberger H. Practical handbook of material flow analysis. Boca Raton: Lewis Publishers; 2004. Gunesegaram D, Givord M, O’Donnell RG, Finnin BR. ATM high pressure die casting and its benefits. In: Proceedings of the 111th NADCA metalcasting congress; 2007. Ramakrishnan S, Tharumarajah A, Koltun P, Roberts MJ. Eco-efficient light-metals component manufacturing. In: Proceedings of the first international light metal technology conference; 2003. p. 125–30. Sustainability Victoria, Resource efficiency measurement—guidance notes, http://www.sustainability.vic.gov.au/resources/documents/2-SV Resource Efficiency guidance notes1.doc, 21 January 2008. Tharumarajah A, Koltun P, Ramakrishnan S, Roberts M. Improving the environmental performance of aluminium die casting production. In: 9th international conference on manufacturing excellence (ICME 2003); 2003. Young K, Eisen P. New die casting technologies—markets and applications. In: Die casting 2000 conference; 2000.
This study assessed the material and cost efficiencies of aluminium and operating resources used to make HPDC parts. In particular, the hidden costs of aluminium losses in the manufacture of HPDC parts. It also, computed the intensity of use and costs for high-use operating resources, viz. electricity, natural gas, die-lube and water. Results of this assessment clearly indicate opportunities for improving the material efficiency of aluminium, thus reducing the costs of losses. It is also evident from the analysis the highuse-dependence of the considered operating resources on the aluminium processed through the plant. Reducing aluminium
References
Contents lists available at ScienceDirect
Resources, Conservation and Recycling journal homepage: www.elsevier.com/locate/resconrec
Benchmarking aluminium die casting operations A. Tharumarajah ∗ CSIRO Sustainable Ecosystems, 37 Graham Road, Highett, Victoria 3190, Australia
a r t i c l e
i n f o
Article history: Received 4 March 2008 Received in revised form 10 June 2008 Accepted 17 June 2008 Available online 31 July 2008 Keywords: Die casting Aluminium Benchmarking Recycling Environment
a b s t r a c t Increasing demand in world automotive markets for aluminium die cast components is creating significant opportunities and challenges for the Australian industry, which is positioning itself as a global player. To meet these challenges, the industry is continuously seeking to improve its overall resource efficiency that can result in the reduction of cost and impact on green house gas (GHG) emissions. In order to understand and evaluate the current position, this study benchmarks the use of aluminium and high-use operating resources of a large representative aluminium high-pressure die casting (HPDC) facility. By modelling the complex web of product, recycling and waste flows, resource efficiencies, costs and GHG impacts of the considered resources are computed. The central focus of this study is the in-house recycling flows of aluminium, normally considered good practice, though it can have negative implications on resource efficiency, costs and GHG impact. In fact, as revealed by this study, the recycling losses contribute as much as around 49% of the total aluminium melted adding about 44% to the cost of manufacture and 50% of the GHG added in production. Using the insights obtained, the technological and other systemic factors that contribute to the losses are identified and areas of improvements are suggested. Crown Copyright © 2008 Published by Elsevier B.V. All rights reserved.
1. Introduction The attractiveness of die casting is its ability to make near net-shape parts with tight tolerances and requiring little or no machining. The more applied of the different die casting processes is high-pressure die casting (HPDC) with high rates of production. The automotive industry uses an extensive range of aluminium HPDC parts including transmission housings, cylinder heads, inlet manifolds, engine sumps as well as decorative trim. This trend is increasing as replacement of steel parts with lighter aluminium HPDC parts grows (Young and Eisen, 2000). To meet the challenges of competition, the industry is continuously seeking to improve its overall performance. While costs, delivery and quality have been the traditional determinants of this, growing concerns about environmental burdens (green house gas (GHG) emissions) created by manufacturing has the industry seeking to reduce the GHG emissions as well. Notably, HPDC processes can be energy and water intensive with high percentage of losses of high cost aluminium as trim wastes. In order to benchmark the flows and wastes, this study considers the metabolism of physical flows in a representative die casting industry that has an estimated 10% of the Australian market. The study focuses on top five of the resources used, namely aluminium
and associated operating resources, viz. water, electricity, gas and die-lubricant (die-lube). Using input–output mapping, the product, waste and circulating (recycling) flows are mapped and the amounts are estimated using material flow analysis (MFA, Brunner and Rechberger, 2004). The estimation is accompanied by calculation of direct and added costs, and the GHG impacts. The central focus being the in-house recycling flows of aluminium. Normally, such recycling is considered good practice, though it can have negative implications on resource efficiency, costs and GHG impact. This estimation reveals the extent of the resource efficiencies that are further examined to investigate the influencing factors. These factors, normally restricted to the die casting process, are expanded to include social factors such as training, maintenance, quality assurance, and shop-floor information and management systems. In Section 2, analysis of the physical flows and their estimation is given. This is followed by the analysis of material efficiencies, costs and environmental impacts from these flows. Section 4 discusses the results of benchmarking and examines technological and other factors that influence the losses and suggests avenues for improvement. Finally, conclusions are drawn with some pointers to future work. 2. Resource flows in HPDC
∗ Tel.: +61 3 9252 6458; fax: +61 3 9252 6249. E-mail address: [email protected].
Investigating usage and cost efficiencies (or intensities) of aluminium (alloy ADC12) and associated operating inputs requires a
0921-3449/$ – see front matter. Crown Copyright © 2008 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.resconrec.2008.06.007
1186
A. Tharumarajah / Resources, Conservation and Recycling 52 (2008) 1185–1189
study of the flows of these through the various operations in the die casting plant. Operating resources, in particular, can be the numerous lubricants (such as die-lubes, tip lubes), chemicals (anti-rust, detergents, silver grease), water, fuels (for maintenance), natural gas, electricity and so on. In this study, the flow of resources is limited to aluminium and operating inputs of electricity, natural gas, die-lube and water used in the production. Usage of these operating resources is largely dictated by the parts produced. To understand the flows, Fig. 1 provides a schematic of the flow of aluminium through the plant and the dependent flow of the operating resources. The flow of aluminium through the plant can be described by the following distinct stages: • Aluminium for production comes from three sources: as hot metal deliveries (HMD), recycled process scrap (in-house, from rejects and process losses) and ADC12 ingots that are input to recycled aluminium. • Melting and holding furnaces store the molten HMD and recycled aluminium. • Parts are cast at the HPDC machines of different tonnages (800 T × 5, 1250 T × 4, 2250 T × 2 and 2500 T × 2) depending on their design with some capability overlap between the machines. • Excess aluminium in parts is trimmed and the part is finished (some parts undergo machining) and the parts, after testing and inspection, are shipped to the customers. The in- and out-flows of aluminium and investigated operating inputs, shown in Fig. 1 can be complex. In particular, a number of waste streams (i.e. non-productive or not saleable product outputs) of aluminium and operating inputs occur right throughout the process. Significant of these (for aluminium) include: • Oxidation losses at melting and holding furnaces (sold to aluminium recyclers). • Process losses (such as warm-ups, mis-runs and others) at the casting process (mostly recycled).
Table 1 HPDC Al flows by weight for the data period Category
Weight (tonnes)
Aluminium processed Aluminium waste sold Aluminium waste recycled in-house Aluminium parts shipped
15,611 507 7,663 7,441
• Yield losses (such as runners, biscuits and flashings, mainly due to die design) that is trimmed (recycled). • Turnings from machining (sold to aluminium recyclers). Wastes from operating inputs include waste water (mixed with die-lube waste) that is treated and disposed as trade waste. Energy losses from electricity and gas use are not captured. Using material flow analysis the inputs and outputs at the various production stages and at supporting processes are mapped. However, the depth of detailed mapping differs due to data availability and the purpose for which analysis is undertaken. For instance, flows of aluminium is tracked at the part level (due to part-dependent differences in shot weight, yield and process losses, and general availability of data), whereas, electricity is categorised by consumption by one or the other processes or equipment. 3. Usage and cost efficiencies 3.1. Aluminium (HPDC) 3.1.1. Usage efficiency The flow (by weight) of aluminium used in HPDC is computed from design data for the individual parts and daily production data from April 2006 through March 2007 for the machine groups. This data is used in computing the material efficiency. Normal efficiencies consider the material in a product and associated waste (that leaves the plant). This efficiency, called here as 1st pass material efficiency, is calculated from the flow data (see
Fig. 1. Flow of aluminium and others through HPDC process.
A. Tharumarajah / Resources, Conservation and Recycling 52 (2008) 1185–1189 Table 2 Cost breakdown of aluminium losses
Table 1) as follows Ist pass material efficiency =
Al-Sh × 100% = 93.87% Al-Sh + Al-Ws
where, Al-Sh is the aluminium parts shipped and Al-Ws is the aluminium waste sold (furnace losses, oily scrap and machining losses). This efficiency, calculated using data in Table 1, indicates that a very low percentage of ∼6.1% of the product material ends up as waste. But, this is not the case since the HPDC process produces a significant amount of hidden aluminium losses, i.e. in-house recycled aluminium that circulates. These losses, nevertheless can contribute substantially to the cost of making the products. Thus, a better measure is to consider the material efficiency from the inputs of aluminium including that which is recycled in-house from process losses. This operational material efficiency is calculated as operational material efficiency =
1187
Al-Mt − Al-Ws − Al-Wr × 100% = 47.7% Al-Mt
where, Al-Mt is the aluminium processed (i.e. total melt), Al-Ws is aluminium waste sold (furnace losses, oily scrap and machining losses), Al-Wr is aluminium waste recycled in-house (yield and process losses) and Al-Sh is the aluminium parts shipped. Comparing the material efficiencies with and without in-house recycling reveals that while only 6.1% of the product leaves the plant as waste (1st pass efficiency), a substantial percentage of 52.3% of the aluminium processed ends up as losses when operational material efficiency is considered. Comparing more detailed operational material efficiencies among machine groups, the highest is for 2500 T (∼52%) and the lowest is for 800 T (∼42%). The lower values for 800 T are mainly due to relatively lower process losses (warm-ups, etc.), though the yield losses (influenced by the part and die design) are higher. Furnace loss is assumed constant at 2.5%. 3.1.2. Cost efficiency For the purpose of computing cost efficiency, waste includes all resources which are non-productive or not saleable as a product (Sustainability Victoria, 2008). That is, it includes all the aluminium losses that occur whether they are recycled in-house or sold to outside recyclers. There are two types of costs associated with aluminium waste (both recycled and sold): • Added cost of waste: production costs incurred in processing aluminium through the plant before waste occurs. For example, cost of melting for oxidation losses or cost of casting before warm-up loses. This cost is estimated for each stage of production from the total cost of manufacture, i.e. cost of materials (other than aluminium), direct labour, other variable (includes other materials, consumables, etc.), indirect labour, depreciation and others. • Cost of handling and transport of waste: these costs are associated with waste sold and are estimated from actual costs.
Loss category
% of total cost of losses
Process losses Yield losses Waste sold Cost of additional Al assigned to waste Total cost of losses
29.8 66.9 0.6 3.3 100.0
3.2. Usage intensities of operating resources The usage intensity of operating resources, viz. electricity, gas, die-lube and water is estimated by category of use. Intensity measure used is consumption per tonne (or kg) of aluminium product sold. In order to allocate their usage for production activities, the category of uses are further analysed and an estimate is made for apportioning to production. Consider the case of electricity, Fig. 2 shows the major categories of its use by consumption and percentage. The breakdown of the intensity of electricity consumption reveals, approximately 40% is consumed in direct production (HPDC machines, CNC machining centres and heat treatment) and a higher percentage of about 47% on production services (air compression, cooling system and 95% lighting). That is, a total of 87% of electricity consumption is in production. Table 3 indicates the usage of all the resources considered in this study. It shows percentage allocated to production, intensity of use per tonne of aluminium sold and % cost added to production. Reviewing the intensities, the intensities for electricity and water are quite high, though without a comparative value it is difficult to judge efficiency of usage. Nevertheless, use of around 4.3 l of (drinking) water per kg of aluminium sold is of concern. Taking the analysis a step further, respectively, around 0.6 l and 1.6 l of water
1 − total cost of losses × 100 = 55.9% total manufacturing costs
A breakdown of the cost of waste is given in Table 2. Cost percentage attributed to the losses takes into account the production stage at which the losses occur. For instance, processes losses (i.e.
13.1 29.5 0.25 1.46 44.1
losses due to warm-up and mis-runs, and other rejects during casting) occur at the casting stage, and hence the added cost of these losses includes cost of melting, holding and casting, and transport and handling. Similarly, yield losses are at the trimming stage. As for waste that is sold, the losses include furnace oxidation losses, oily scrap at HPDC and machining losses. Waste sold generates revenue, but the net result is a cost to plant. The highest source of added costs of waste is yield losses, which largely depends on the part and die design, and thus can be difficult to reduce in the short-term. On the other hand, there is potential to reduce process losses that amount to approximately 30% of the cost of all losses or about 13% of the manufacturing cost. The cost of additional aluminium to compensate for the shortfall (aluminium parts and waste sold are greater than the aluminium recycled inhouse) is small at 1.46% of manufacturing cost.
Resource cost efficiency of aluminium is computed using estimated cost of losses (shown as a % of manufacturing cost in Table 2) as follows resource cost efficiency =
% of total cost of manufacture
Fig. 2. Electricity consumption by category of use.
1188
A. Tharumarajah / Resources, Conservation and Recycling 52 (2008) 1185–1189
Table 3 Usage and cost of operating resources
and yield losses. Thus, a fair proportion of around 49% of the total GHG is added for the losses.
Operating resource
% allocated to production
Usage intensity/ tonne of Al sold
% of manufacturing cost
Electricity (kWh) Natural gas (MJ) Die-lube (L) Water (L)
87.5 100.0 100.0 78.4
1.54 0.02 0.01 4.32
1.33 1.02 0.65 0.11
is expended on process and yield losses (process and yield losses account for around 13% and 36% of total aluminium processed in the plant, respectively). Interesting aspect of the cost operating resources is that they only contribute a small percentage of around 3% to the cost of manufacture with the highest contributor being electricity. 3.3. Environmental impact In computing the environmental impact, i.e. GHG, only contributions by way of additional aluminium bought from outside and usage of operating resources is taken into account. The additional aluminium compensates for the balance between what is bought, recycled, sold as product and sold as waste. Thus, it is considered as part of the waste. Total amount of additional aluminium is around is 224 T or 0.03 kg per kg of aluminium sold by the plant. In terms of GHG impact this translates to around 0.5 kg of CO2 equiv. per kg of aluminium sold, assuming a GHG impact of 16.5 kg CO2 -equiv. per kg (this assumes a composition of 50:50 primary to secondary aluminium produced in Australia, Ramakrishnan et al., 2003). The second source of added GHG is the operating resources considered in this study. Based on the usage intensities of these resources (see Table 3) GHG impact in terms of kg CO2 -equiv. for each of the operating resources is calculated. As shown in Table 4, of the ∼4.1 kg CO2 -equiv. of GHG added in producing 1 kg of finished casting, around 88% is contributed by the operating resources, with electricity and water contributing around 51% and 37%, respectively. The higher impact of electricity is due to the use of brown coal that has a very high global warming potential for generating electricity. Aside, the cost of these two resources is a fraction of total manufacturing cost (Table 3), and hence incentive to reduce these comes from their environmental impact, rather than cost. In the case of water, water used in production does not have a high GHG burden. The burden is added in treatment with impacts coming from use of caustic soda, etc. (Tharumarajah et al., 2003). Contribution from natural gas is negligible due to its lower impact and less usage intensity and so is die-lube. The table also shows the portion of the added GHG that can be attributed to process and yield losses that are recycled. Assuming a linear proportion, the impact due to these can be calculated, i.e., respectively, 13% and 36% of the total GHG are attributed to process Table 4 Greenhouse gas impact of operating resources Item
GHG (kg CO2 -equiv.) Per kg of Al sold
Allocated to recycled process losses per kg Al sold
Allocated to recycled yield losses per kg sold
Additional Al Electricity Natural gas Die-lube Water
0.50 2.15 0.00 0.01 1.54
0.07 0.29 0.00 0.00 0.21
0.18 0.78 0.00 0.00 0.56
Total
4.19
0.56
1.52
4. Discussion on improvement The estimation of aluminium and operating resources reveals the extent to which aluminium losses contribute to costs and GHG impacts. While most of these losses are recycled, nevertheless, the hidden cost of these losses can be high (see Table 2) and lessening of these losses can reduce the waste (sold). Additionally, the highusage operating resources (electricity, gas, water and die-lube) considered here have very high dependence on the aluminium processed. Thus, savings in aluminium losses is bound to save on these too. In improving resource efficiency and reducing cost of losses (see Tables 1 and 2), two principal avenues can be considered. Yield losses occur mainly due to excess metal in runners and flashings. These losses depend on the part and die design. However, recent technological advances based on novel runner design and semisolid feed system, such as advanced thixotropic metallurgy (ATM, Gunesegaram et al., 2007), can reduce the shot weight and improve porosity of cast parts. Application and acceptance of such technologies require extensive industry trials before becoming main stream technologies and hence is considered to be a medium–long term opportunity. On the other hand, process losses are a result of casting machines being unable to maintain operating conditions during production, such as warm-up and mis-runs. It can also be due to machine maintenance. Improvements in this area offer a viable short-term option. Discussion on improvements is thus focussed on factors that underpin these losses. The two most causes of process losses are warm-ups and misruns. Warm-up losses occur due to the need to warm-up the die by using the heat from molten aluminium during start up phase of producing a part. Warm-up castings are rejects and are recycled. The number of warm-up castings before acceptable parts are produced depends on many factors including the condition of the machine to maintain an optimum temperature and pressure (to inject the molten metal in to the die cavity). It can also be affected by the concentration and amount of warm-up release agent used when the die is cold. A mis-run or cold shut, on the other hand, occurs when a casting lacks completeness due to the failure of the metal to fill the die cavity. This may be due to poor fluidity of the metal, machine unable to maintain temperature and pressure optimum for casting or due to part and die features such as too thin sections and wall thickness and improper gating system. Thus, the primary reason of these losses can be attributed to the condition of the casting process and/or the part and die design. To add to these issues, the nature of the casting process is such that when disruptions to product cycle are experienced, the optimum conditions tend to deteriorate. These instances include breakdown, die-changes, process cleanup and others. Thus, there may be other secondary issues other than those with machines, dies and quality of aluminium. In fact, the secondary issues can be much deeper, such as inadequate operator training, lack of timely information and delays in reacting to problems, and lack of knowledge of processes and idiosyncrasies of parts, in particular new parts. Thus, as shown in Fig. 3, they become influential in trying to improve product quality, though more often than not improvements are directed at dealing with problems at the machines, dies or melt quality. The danger of such myopic approaches is that the secondary issues can fester and tend to become systemic affecting overall productivity of the shop-floor.
A. Tharumarajah / Resources, Conservation and Recycling 52 (2008) 1185–1189
1189
losses can have a positive flow-on effect in reducing the consumption of these too. Additional benefits may arise in the reduction of greenhouse gases, in particular through the reduction of electricity consumption. In essence, a study to benchmark flows can reveal many insights when circulating in-plant recycling flows are considered. In this case, in-house aluminium recycling though considered a benefit to the operation, nevertheless adds substantially to the cost and GHG emissions. In fact, quantification of these impacts can be an eye opener in exposing the inefficiencies which is otherwise hidden, in particular GHG contributions of low cost resources. The study also briefly examined process and softer systemic factors for reducing the losses. Some of the latter, such as operator knowledge can be critical, in particular when quality is affected by the complexity and idiosyncrasies of the parts produced. Others, such as shop-floor management and information systems also may act as impediments when quick diagnosis and action are required. Fig. 3. Primary and secondary factors influencing part quality and shop-floor efficiency.
Acknowledgements
Other than savings in aluminium, opportunities exist for separation, recovery and reuse of die-lube from waste water. The analysis has revealed the savings in pursuing this avenue. Such recycling along with rain water harvesting could lead to self-sufficiency in water.
The author wishes to thank Sustainability Victoria and Cooperative Research Centre for Casting and Solidification Technologies for their support in conducting this study.
5. Conclusion
Brunner PH, Rechberger H. Practical handbook of material flow analysis. Boca Raton: Lewis Publishers; 2004. Gunesegaram D, Givord M, O’Donnell RG, Finnin BR. ATM high pressure die casting and its benefits. In: Proceedings of the 111th NADCA metalcasting congress; 2007. Ramakrishnan S, Tharumarajah A, Koltun P, Roberts MJ. Eco-efficient light-metals component manufacturing. In: Proceedings of the first international light metal technology conference; 2003. p. 125–30. Sustainability Victoria, Resource efficiency measurement—guidance notes, http://www.sustainability.vic.gov.au/resources/documents/2-SV Resource Efficiency guidance notes1.doc, 21 January 2008. Tharumarajah A, Koltun P, Ramakrishnan S, Roberts M. Improving the environmental performance of aluminium die casting production. In: 9th international conference on manufacturing excellence (ICME 2003); 2003. Young K, Eisen P. New die casting technologies—markets and applications. In: Die casting 2000 conference; 2000.
This study assessed the material and cost efficiencies of aluminium and operating resources used to make HPDC parts. In particular, the hidden costs of aluminium losses in the manufacture of HPDC parts. It also, computed the intensity of use and costs for high-use operating resources, viz. electricity, natural gas, die-lube and water. Results of this assessment clearly indicate opportunities for improving the material efficiency of aluminium, thus reducing the costs of losses. It is also evident from the analysis the highuse-dependence of the considered operating resources on the aluminium processed through the plant. Reducing aluminium
References