As the foundation of modern manufacturing, the casting industry continues to face substantial challenges in energy efficiency, resource consumption, and environmental impact. Sand casting, as the most widely used casting method, is particularly significant in this regard. In this work, I focus on the process design features that directly influence product quality in sand casting, and classify these features into material features, geometric structure features, and pouring features. By establishing a carbon benefit model that integrates carbon emissions, production time, and resource costs, I aim to provide a comprehensive evaluation framework for sand casting parts. The proposed model enables the optimization of process parameters associated with design features during the design stage, thereby improving the sustainability and efficiency of sand casting production. This study demonstrates the effectiveness of the carbon benefit model through two industrial case studies: a disk-shaped casting and a base casting. In the first case, the geometric structure features related to hole formation are optimized by comparing machining and sand core methods. In the second case, the pouring features such as sprue design, gate cross-section ratio, and riser parameters are optimized to eliminate internal defects while improving carbon benefit. The results show that the optimization of process design feature parameters can significantly reduce carbon emissions, shorten production time, and lower resource costs, while maintaining the required casting quality. The proposed carbon benefit model offers a practical tool for foundry engineers to select more sustainable manufacturing strategies for sand casting parts.
1. Introduction
The traditional manufacturing sector consumes more than 80% of the world’s total industrial energy, and its share continues to increase with ongoing industrialization. In 2018, China’s manufacturing industry accounted for approximately 55.65% of the country’s total energy consumption. This massive energy consumption makes manufacturing a major contributor to greenhouse gas emissions. Among all manufacturing sectors, the foundry industry is especially energy-intensive. China has been the world’s largest producer of castings for many years, producing about 47.2 million tonnes in 2016, which represented roughly 45% of global casting output. However, the energy consumption per tonne of castings in China is still about twice that of developed countries. Moreover, unreasonable process design and redundant production steps lead to lower yield and unnecessary energy consumption. Therefore, reducing energy use and environmental impact in the foundry industry is an urgent and meaningful task.
Previous research has addressed energy saving and emission reduction in manufacturing from various perspectives, such as life cycle assessment, energy benchmarking, and process parameter optimization. Some studies have focused on carbon emission modeling and optimization for machining processes, while others have investigated casting process design parameters. However, few studies have considered the comprehensive benefit that combines carbon emissions, production time, and resource costs for sand casting parts. A notable gap exists in the evaluation of process design features at the product design stage, where the influence of design features on downstream processes can be quantified and optimized.
In this research, I propose a carbon benefit model specifically for sand casting process design features. The model quantifies carbon emissions, production time, and resource costs as functions of three categories of process design features: casting material features, casting geometric structure features, and pouring features. Based on this model, an optimization method is developed to adjust the associated process parameters in order to improve the carbon benefit while satisfying quality constraints. Two case studies are presented to verify the applicability of the proposed method.
2. Analysis and Classification of Sand Casting Process Design Features
Sand casting is a versatile manufacturing process capable of producing components ranging from a few grams to hundreds of tonnes. The typical process route for sand casting parts includes raw material melting, alloying, molding, core making, pouring, solidification, cooling, shakeout, fettling, and inspection. Each step consumes energy and materials and generates waste and emissions. Figure 1 illustrates the overall process flow of sand casting production.
The process design features of sand casting can be categorized into three main groups according to their role in determining the quality of sand casting parts. I define these categories as follows:
Definition 1 (Casting Material Features, CF): Features related to material properties in the casting process, such as molding sand type, thermal diffusivity of the mold, alloy type, and hardener ratio.
Definition 2 (Casting Geometric Structure Features, HF): Features related to the geometric configuration of the casting and the mold, such as wall thickness, cast holes, fillets, grooves, reinforcing ribs, mold wall thickness, and core count.
Definition 3 (Pouring Features, ZF): Features related to the gating and feeding system, such as runner type, runner dimensions, gate cross-section ratio, riser type, and riser dimensions.
These features are not independent. They interact with the selected technological parameters and significantly influence the resource consumption, production time, and emission profile of the entire manufacturing process. For instance, the choice of molding sand material determines the binder system, compaction method, and recycling strategy. The presence of holes or pockets in a casting may require cores, which in turn affect the mold assembly time and the amount of metal poured. Pouring features such as gate design and riser size determine the filling behavior and feeding efficiency, which influence defect formation and casting yield.
3. Carbon Benefit Model for Sand Casting Process Design Features
3.1 Total Carbon Emission Model
Carbon emissions from the sand casting process can be classified into three sources: energy-related emissions, material-related emissions, and process-related emissions. Energy-related emissions arise from the consumption of electricity, coke, natural gas, and other fuels in various production departments. Material-related emissions originate from the production, transportation, and processing of raw materials such as pig iron, steel scrap, sand, coatings, and water. Process-related emissions are directly released into the atmosphere during chemical reactions, such as the decomposition of organic binders in resin sand and the calcination of limestone in the melting process.
The total carbon emission associated with process design features is expressed as:
$$CE_{Total} = C_N + C_M + C_P \tag{1}$$
where \(C_N\) is the energy-related carbon emission, \(C_M\) is the material-related carbon emission, and \(C_P\) is the process-related carbon emission. Each component is further decomposed based on the three categories of process design features. Energy-related emissions are calculated as:
$$C_N = \sum_{i=1}^{u} m(a_{CF}^{i}) \cdot EF_i + \sum_{j=1}^{v} m(a_{HF}^{j}) \cdot EF_j + \sum_{k=1}^{w} m(a_{ZF}^{k}) \cdot EF_k \tag{2}$$
where \(m(a_{CF}^{i})\) denotes the consumption of the \(i\)-th energy carrier linked to casting material features, and \(EF_i\) its carbon emission factor. The superscripts \(HF\) and \(ZF\) refer to geometric structure features and pouring features, respectively. To unify different energy types, the standard coal equivalent is used:
$$m(A) = a_A \cdot e_A \tag{3}$$
where \(a_A\) is the physical amount of energy \(A\) and \(e_A\) is the standard coal conversion factor.
Material-related emissions are given by:
$$C_M = \sum_{r=1}^{o} m(b_{CF}^{r}) \cdot f_r + \sum_{s=1}^{p} m(b_{HF}^{s}) \cdot f_s + \sum_{t=1}^{q} m(b_{ZF}^{t}) \cdot f_t \tag{4}$$
where \(m(b_{CF}^{r})\) is the consumption of the \(r\)-th material associated with casting material features, and \(f_r\) is its carbon emission factor.
Process-related emissions are expressed as:
$$C_P = \sum_{d=1}^{x} V(c_{CF}^{d}) \cdot a_d + \sum_{e=1}^{y} V(c_{HF}^{e}) \cdot a_e + \sum_{f=1}^{z} V(c_{ZF}^{f}) \cdot a_f \tag{5}$$
where \(V(c_{CF}^{d})\) is the volume of gas generated by the \(d\)-th chemical reaction associated with casting material features, and \(a_d\) is its carbon emission factor.
3.2 Production Time Model
The total production time for sand casting parts is calculated from the start of sand mixing to the final inspection. It includes both the main processing time of major equipment and auxiliary time such as loading, unloading, and material handling. The production time affected by process design features is formulated as:
$$PT_{Total} = \sum_{i’=1}^{m} \sum_{j’=1}^{n} \left[ \bar{t}_{CF}^{i’j’}(N) + \bar{t}_{HF}^{i’j’}(N) + \bar{t}_{ZF}^{i’j’}(N) \right] + \sum_{i’=1}^{m} \sum_{k’=1}^{n’} \left[ \bar{t}_{CF}^{j’k’}(N) + \bar{t}_{HF}^{j’k’}(N) + \bar{t}_{ZF}^{j’k’}(N) \right] \tag{6}$$
where \(\bar{t}_{CF}^{i’j’}(N)\) is the average main processing time for the \(j’\)-th operation associated with casting material features in the \(i’\)-th production department, and \(\bar{t}_{CF}^{j’k’}(N)\) is the average auxiliary time. \(N\) denotes the batch size of sand casting parts. When multiple machines operate in parallel, the longest processing time is used.
3.3 Resource Cost Model
Resource cost includes energy cost, material cost, labor cost, and residual cost. Energy cost covers electricity and other fuels. Material cost includes both primary materials (pig iron, steel scrap, returns) and auxiliary materials (resin, coating, water, shot). Labor cost is computed based on worker hours, and residual cost includes expenses for tooling wear, depreciation, and consumables. The total resource cost is:
$$RC_{Total} = RC_{energy} + RC_{material} + RC_{labor} + RC_{rest} \tag{7}$$
According to the feature classification, each cost type can be further decomposed as:
$$RC_X = RC(d_{CF}) + RC(d_{HF}) + RC(d_{ZF}), \quad X \in \{energy, material, labor, rest\} \tag{8}$$
3.4 Carbon Benefit Definition
To evaluate the combined performance of carbon emissions, production time, and resource costs, I introduce the carbon benefit index \(TC\) as:
$$TC = \alpha \cdot CE_{Total} + \beta \cdot PT_{Total} + \mu \cdot RC_{Total} \tag{9}$$
where \(\alpha\), \(\beta\), and \(\mu\) are weighting factors that reflect the priorities of the foundry. A lower \(TC\) value indicates better overall benefit. The weighting factors are determined based on managerial goals; for example, if emission reduction is the top priority, \(\alpha\) is set higher.
4. Parameter Optimization of Sand Casting Process Design Features
4.1 Design of Associated Process Parameters
The parameters linked to casting material features, such as alloy type and molding sand composition, are usually fixed by the foundry’s conditions. Therefore, I focus on the parameters associated with casting geometric structure features and pouring features. Table 1 shows the recommended machining allowance grades for gray iron castings produced by machine molding.
| Maximum dimension (mm) | Grade D | Grade E | Grade F | Grade G |
|---|---|---|---|---|
| 160–250 | 1.1 | 1.4 | 2.0 | 2.8 |
| 250–400 | 1.4 | 1.4 | 2.5 | 3.5 |
| 400–630 | 1.6 | 2.2 | 3.0 | 4.0 |
For the minimum cast hole diameter, Table 2 provides reference values based on production batch.
| Production batch | Gray iron (mm) | Cast steel (mm) |
|---|---|---|
| Mass production | 12–15 | — |
| Batch production | 15–30 | 30–50 |
| Single/small batch | 30–50 | 50 |
The pouring time can be estimated using the empirical formula:
$$\tau = S_1 \sqrt{m} \tag{10}$$
where \(S_1\) is a coefficient dependent on wall thickness and \(m\) is the total mass of poured metal. The pouring temperature depends on the alloy and the main wall thickness. Table 3 presents recommended pouring temperatures for HT250 iron.
| Main wall thickness (mm) | Tapping temperature (°C) | Pouring temperature (°C) |
|---|---|---|
| 8–15 | 1470 | 1390–1450 |
| 15–30 | 1460 | 1370–1440 |
| 30–50 | 1430 | 1350–1430 |
| >50 | 1420 | 1270–1360 |
4.2 Optimization Objective Function
The optimization problem for process design feature parameters is formulated as:
$$\begin{aligned}
\text{min} \quad & TC = \alpha \cdot CE_{Total} + \beta \cdot PT_{Total} + \mu \cdot RC_{Total} \\
\text{s.t.} \quad & R_1 = \{X \mid g_i(X) \ll K_1\} \\
& R_2 = \{X \mid m_i(X) \ll K_2\} \\
& T_{min} \le T \le T_{max} \\
& cond_1, \dots, cond_m \mid b_{min} \le V \le cond_1, \dots, cond_m \mid b_{max}
\end{aligned} \tag{11}$$
Here, \(K_1\) represents the total shrinkage porosity limit and \(K_2\) represents the void fraction limit obtained from casting simulation. \(T\) is the pouring temperature, \(V\) represents the pouring velocity, and \(cond_1,\dots,cond_m\) denote relevant process conditions such as wall thickness and gating system dimensions. The quality constraints \(R_1\) and \(R_2\) are validated through simulation using ProCAST software.
4.3 Low-Carbon Optimization Procedure
The optimization steps are as follows:
- Selection of feature-related parameters. Extract the process parameters associated with the studied process design features and set initial values based on handbooks and foundry conditions.
- Creation and quality validation of the original process plan. Build the original casting process plan and run pouring simulation in ProCAST. Confirm that the total shrinkage porosity and void fraction meet the quality limits.
- Calculation of carbon benefit. For all feasible plans, calculate carbon emissions, production time, and resource costs, and then compute the carbon benefit \(TC\).
- Determination of the optimized plan. Modify the process parameters, re-run simulation, recalculate \(TC\), and select the plan with the lowest carbon benefit as the optimized solution.
5. Case Studies and Application Analysis
5.1 Case 1: Geometric Structure Feature of a Disk Casting
The first case involves a disk-shaped gray iron casting (HT250) with a net weight of 186.138 kg. The casting has three holes of 45 mm diameter, which require high quality. Two alternative process routes are compared: Plan FE1 manufactures the holes by machining after casting, while Plan FE2 forms the holes using sand cores during molding. A third optimized plan FE3 is derived from FE2 by adjusting the hole machining allowance and core diameter. Table 4 lists the key process parameters for the three plans.
| Parameter | FE1 | FE2 | FE3 |
|---|---|---|---|
| Pattern draft (°) | 1.6 | 1.6 | 1.6 |
| Pouring temperature (°C) | 1420 | 1420 | 1420 |
| Gas evolution (ml/g) | 15.5 | 15.5 | 15.5 |
| Pouring time (s) | 6 | 6 | 6 |
| Cooling time (h) | 24 | 24 | 24 |
| Surface machining allowance (mm) | 2.5 | 2.5 | 2.5 |
| Hole machining allowance (mm) | 0 | 1.5 | 2.5 |
| Core diameter (mm) | 0 | 42 | 40 |
The pouring simulation results show that all three plans satisfy the quality requirements regarding total shrinkage porosity and void fraction. The carbon emission details are calculated using the models in Section 3. Table 5 summarizes the main emission items.
| Emission source | FE1 | FE2 | FE3 |
|---|---|---|---|
| Pig iron | 646.405 | 634.038 | 633.941 |
| Steel scrap | 4244.098 | 4162.904 | 4162.160 |
| Silica sand | 154.063 | 156.142 | 155.839 |
| Steel shot | 15.798 | 15.948 | 15.918 |
| Water | 0.349 | 0.349 | 0.349 |
| Auxiliary materials | 81.396 | 89.942 | 81.239 |
| Coating | 90.902 | 91.752 | 91.603 |
| Alcohol | 3.390 | 3.390 | 3.390 |
| Natural gas | 34.880 | 34.880 | 34.880 |
| Coke | 195.250 | 191.515 | 194.442 |
| Electricity (equipment) | 397.351 | 377.391 | 387.670 |
| Melting energy | 1043.876 | 1023.905 | 1025.941 |
| Core making energy | 0 | 39.982 | 27.188 |
| Machining energy | 20.900 | 2.300 | 4.800 |
| Process emissions | 24.699 | 25.150 | 25.024 |
From Table 5, the steel scrap emission dominates the total carbon footprint. FE1 uses more molten metal because the holes are machined after casting, leading to a larger poured weight. FE2 and FE3, which use cores, reduce the metal poured and thus lower emissions. FE3, with optimized hole allowance and core diameter, achieves the lowest total carbon emission. The comparison of total carbon emissions is shown in Figure 2.
The production time for casting 10 pieces is calculated using Equation (6). The machining plan (FE1) requires extra machining time for drilling the three holes, while the core-making plans eliminate this step. As a result, FE2 reduces production time by 1.353% compared to FE1, and FE3 by 1.454%. FE3 also has the shortest sand treatment time due to the smaller core size.
Resource cost calculations include material costs, energy costs, labor costs, and residual costs. FE1 has higher material cost because of increased metal melting, and higher labor cost due to machining operations. FE2 reduces total cost by 16.084% compared to FE1, and FE3 reduces it by 16.503%. The lowest cost is achieved by FE3.
Using weighting factors \(\alpha = 0.45\), \(\beta = 0.78\), and \(\mu = 0.62\), the carbon benefit values are calculated as follows:
- FE1: \(TC = 12012.703\)
- FE2: \(TC = 10548.009\)
- FE3: \(TC = 10503.226\)
The results indicate that using sand cores for hole formation provides a significantly better carbon benefit than machining. Further optimization of the geometric structure feature parameters in FE3 results in the best overall performance among the three plans.
5.2 Case 2: Pouring Feature of a Base Casting
The second case study focuses on a base casting used for road traffic barriers. The casting has a maximum dimension of 400 mm × 300 mm × 182 mm and a net mass of 69.882 kg, made of HT250. A semi-closed gating system is initially designed with a cross-section ratio of \(A_{\text{runner}} : A_{\text{sprue}} : A_{\text{gate}} = 1.256 : 1.4 : 1\). Three process plans are considered: FE1 (initial design), FE2 (initial design with a shrink-neck riser to eliminate internal defects), and FE3 (optimized design with modified pouring temperature, pouring time, gate ratio, and a smaller riser). The process parameters are listed in Table 6.
| Parameter | FE1 | FE2 | FE3 |
|---|---|---|---|
| Pattern draft (°) | 0.4 | 0.4 | 0.4 |
| Gas evolution (ml/g) | 15.5 | 15.5 | 15.5 |
| Pouring temperature (°C) | 1420 | 1420 | 1400 |
| Pouring time (s) | 8 | 8 | 6 |
| Cooling time (h) | 24 | 24 | 24 |
| Machining allowance (mm) | 2 | 2 | 2 |
| Cross-section ratio | 1.256:1.4:1 | 1.256:1.4:1 | 1.256:1.3:1 |
| Gating system mass (kg) | 5.727 | 5.727 | 5.271 |
| Riser type | None | Shrink-neck | Shrink-neck |
| Riser quantity | 0 | 1 | 1 |
| Riser mass (kg) | 0 | 0.193 | 1.521 |
| Total feeding system mass (kg) | 5.727 | 5.920 | 6.792 |
The initial plan FE1 shows internal porosity in the simulation, as indicated by purple regions in the total shrinkage porosity plot. To eliminate this defect, a shrink-neck riser is added in FE2, which successfully removes the porosity. FE3 further optimizes the pouring feature by reducing the pouring temperature to 1400 °C, shortening the pouring time to 6 s, and adjusting the gate ratio to 1.256:1.3:1. The optimized plan FE3 also passes the quality requirements.
The carbon emission factors for energy and materials are listed in Table 7.
| Source | Emission factor | Source | Emission factor |
|---|---|---|---|
| Coke | 2.8601 kgCO₂/kgce | Steel scrap | 8.2 kgCO₂/kg |
| Alcohol | 0.480 kgCO₂/kgce | Silica sand | 0.2543 kgCO₂/kg |
| Standard coal | 2.4910 kgCO₂/kgce | Coating | 6.0232 kgCO₂/kg |
| Pig iron | 2.1300 kgCO₂/kg | Water | 1.204 kgCO₂/t |
The material consumption for each plan is calculated based on the total poured mass, including castings, gating systems, and risers. Table 8 shows the quantities of major materials.
| Material | FE1 | FE2 | FE3 |
|---|---|---|---|
| Pig iron (kg) | 11.0389 | 11.1150 | 11.0005 |
| Steel scrap (kg) | 37.6533 | 37.9127 | 37.5223 |
| Resin sand (kg) | 34.6344 | 33.6351 | 33.6620 |
| Coating (kg) | 0.9925 | 0.9520 | 0.9785 |
| Water (kg) | 0.3506 | 0.3531 | 0.3495 |
| Steel shot (kg) | 0.1020 | 0.1020 | 0.1020 |
Table 9 presents the carbon emission details for the three base casting plans.
| Emission source | FE1 | FE2 | FE3 |
|---|---|---|---|
| Pig iron | 23.5129 | 23.6749 | 23.4311 |
| Steel scrap | 308.7569 | 310.8845 | 307.6829 |
| Silica sand | 8.8075 | 8.5534 | 8.5602 |
| Steel shot | 0.8364 | 0.8364 | 0.8364 |
| Water | 0.0004 | 0.0004 | 0.0004 |
| Auxiliary materials | 3.2561 | 3.2528 | 3.2552 |
| Coating | 5.9780 | 5.7341 | 5.8937 |
| Alcohol | 1.3265 | 1.3379 | 1.3227 |
| Coke | 3.5168 | 3.5410 | 3.5045 |
| Electricity | 40.0828 | 45.3363 | 39.8064 |
| Process emissions | 1.2330 | 1.1280 | 1.2160 |
Among the three plans, FE2 has the highest carbon emissions because of the additional riser mass and the unchanged gating system. FE3 achieves a 0.981% reduction in carbon emissions compared to FE2, primarily due to the optimized gating cross-section and lower pouring temperature, which reduces melting energy consumption. Figure 3 compares the total carbon emissions across the three plans.
The production time is also calculated. FE3 has the shortest production time due to the reduced pouring time and optimized gating system. The detailed production time comparison is shown in Figure 4.
The resource costs are calculated based on market prices and labor rates. The worker allocation for the base casting production is given in Table 10.
| Department | Operation | Number of workers |
|---|---|---|
| Melting | Melting | 2 |
| Molding | Jolt-squeeze molding | 1 |
| Stripping | 2 | |
| Flow coating | 1 | |
| Pouring | 3 | |
| Sand treatment | Sand mixing | 2 |
| Shakeout | 2 | |
| Old sand recovery | 2 | |
| Fettling | Removing gate/riser | 1 |
| Machining | 1 | |
| Shot blasting | 1 |
The resource cost comparison shows that FE3 has the lowest cost among the three plans. FE3 reduces total resource cost by 1.384% compared to FE2. With the same weighting factors as in the first case, the carbon benefit values are:
- FE1: \(TC = 314.498\)
- FE2: \(TC = 317.500\)
- FE3: \(TC = 313.211\)
Thus, FE3 achieves the best carbon benefit. The optimization of pouring feature parameters, including the gating cross-section ratio, pouring temperature, pouring time, and riser size, effectively improves the overall sustainability of sand casting parts production.
6. Conclusions and Outlook
In this work, I have developed a comprehensive carbon benefit model for sand casting process design features. The main conclusions are summarized as follows:
- The classification of process design features into material features, geometric structure features, and pouring features provides a clear framework for linking design decisions to carbon emissions, production time, and resource costs.
- The proposed carbon benefit model integrates these three crucial indicators and allows foundry engineers to evaluate the overall performance of alternative process plans for sand casting parts.
- Through the case studies of a disk casting and a base casting, the model successfully identified that the use of sand cores for hole formation yields better carbon benefit than machining, and that optimization of pouring parameters further improves the benefit.
- The combination of casting simulation and carbon benefit optimization enables the design of process plans that are both defect-free and sustainable.
Future research should consider integrating new manufacturing technologies, such as additive manufacturing for pattern and core production, into the sand casting process. Furthermore, the carbon benefit model can be extended to other casting processes and to broader manufacturing systems that involve multiple process chains. By applying the proposed evaluation method, traditional manufacturing industries can move toward higher resource efficiency and lower environmental impact, contributing to global sustainable development goals.