Problem:
Vale, a leading international mining company, needed to optimize traffic flow at one of their mines, where haul trucks and small vehicles caused congestion. Additionally, at another mine, increased train loading operations strained weigh station capacity, requiring mining infrastructure optimization.
Solution:
The company developed purpose-built simulation models in AnyLogic, leveraging its discrete-event and agent-based capabilities. The models incorporated elements from road traffic and process modeling libraries to simulate vehicle movement and interactions, as well as GIS maps for accurate layout representation.
Results:
- The usage of higher-capacity trucks increased the average open time at the gate barrier by 50%.
- The existing road system was found to support operations efficiently, even with a potential 40% increase in traffic.
- Cutting unnecessary weigh-ins and adding a second weighbridge doubled unloading processing capacity.
Introduction: operating worldwide and running ambitious projects
Vale is one of the largest mining companies in the world, operating in 18 countries and specializing in producing iron ore, nickel, and copper. The company runs some of the largest iron ore projects, requiring efficient mine traffic management and infrastructure optimization processes to ensure operational security and productivity.
Use case 1: impact evaluation of an increase in waste-crushing capacity
Problem
An increase in production capacities was proposed at the S11D mine, one of the world's most significant iron ore projects. The new operations would involve haul trucks and light vehicles, which cannot safely share the same road because of their size differences.
Vale needed to evaluate the impact of this operational change on existing traffic, particularly regarding congestion and delay times at the gate barrier. The effects of higher and lower payload trucks on road congestion also required comparison and analysis for decision-making and effective mine traffic management.
highlighting the analyzed location (click to enlarge)
Solution
Vale decided to simulate operations and adopted AnyLogic, the leading simulation modeling software with discrete-event and agent-based simulation capabilities.
Vale chose AnyLogic due to several deciding factors:
- Available useful libraries (such as fluid, road traffic, and process modeling libraries) in the software that facilitate and speed up simulation model creation.
- Integrated GIS maps.
- Possibility of agent reuse in different case scenarios.
Firstly, Vale collected extensive data on vehicle movements, including buses, light cars, and haul trucks. This data provided insights into traffic behavior and congestion patterns. Subsequently, this real-world data was converted into a scheduled rate for the model.
during weekdays and weekends (click to enlarge)
Vale developed an agent-based model with specifically defined patterns to ensure that it followed real-world logic:
- Schedules defined the arrival of both haul trucks and light vehicles.
- The agent controlled the entrance and exit of vehicles in the controlled area.
- Entrance was controlled by setting the TrafficLight on each StopLine before entering the restricted area.
- An additional agent comprised of ParkingLot with a delay for unloading and control of empty positions and queueing was implemented.
with a gate barrier at an intersection and the truck’s parking lot
The study also compared higher-payload haul trucks with lower-payload ones to determine which fleet configuration would most efficiently minimize congestion at the gate barrier.
Results
The simulation results provided critical insights that guided Vale’s decision-making in mine traffic management.
The maximum capacity of the gate barrier was discovered to be 645 trucks per hour for lower payload vehicles and 669 trucks per hour for higher payload vehicles. Thus, using higher-capacity haul trucks reduced the number of trips, leading to an increase of 50% in average open time at the gate barrier.
with lower and higher payload trucks (click to enlarge)
Moreover, even with a potential 20 to 40% increase in traffic demand, the existing road system was found to be capable of supporting operations efficiently. The simulation model gave insights into the potential percentage of impacted vehicles due to traffic and their average time loss, which didn’t exceed 2.66 minutes, meeting the company’s targets for efficiency.
and their average time loss under different traffic scenarios (click to enlarge)
Use case 2: weigh stations optimization
Problem
At another location, the Serra Leste rail loop, additional train loading was proposed to increase Vale’s logistics network utilization. The new loading operations required evaluating the capacity of the existing weigh station and assessing whether the company needed additional weighbridges.
Expanding the current mining infrastructure would come with significant earthworks, making exploring and validating solutions in a risk-free environment essential.
at the Serra Leste rail loop
Solution
Vale developed another simulation model to assess the capacity of the existing mining infrastructure under different scenarios for the weigh station optimization at the Serra Leste rail loop. A sensitivity analysis was conducted on truck arrival rates and the necessity of weighing empty trucks to identify potential bottlenecks. Various alternatives were considered, including relocating the weigh bridge outside restricted areas and reducing unnecessary weigh-ins.
Read also about another project of Vale, focused on productivity improvement of coal mining operations in Moatize, Mozambique.
Results
Simulation results clearly showed the following:
- Reducing empty weighting from 100% to 20% would have a small increase in capacity for the existing bridge.
- Moving the weighbridge outside of the restricted area could provide only a minor improvement in weigh station capacity optimization.
- Moving the weighbridge outside the restricted area and reducing empty weighing could significantly increase capacity.
- Operating two weighbridges outside the restricted area would nearly double the capacity compared to a single weighbridge at 100% weighing.
- Reducing empty weighting with two weighbridges would give only small improvements.
The worst-case scenario was considered 100% (click to enlarge)
By leveraging simulation-driven analysis, Vale successfully optimized its mine traffic management and weigh station operations, ensuring both efficiency and scalability. The presented cases highlight the power of simulation in addressing complex challenges, reinforcing its value in large-scale industrial operations.
The case study was presented by Enrico Miranda from Vale at the AnyLogic Conference 2024.
The slides are available as a PDF.
