PS. Haulage Calculations
- 1 “Instantaneous Trucks” vs “TUM Trucks”
- 2 Truck Assignments
- 2.1.1 Limiting Trucks
- 2.1.1.1 Agents
- 2.1.1.2 Truck Rules
- 2.1.1 Limiting Trucks
- 2.2 Reporting Truck Usage
- 2.2.1 Instantaneous Trucks
- 2.2.1.1 Steady States
- 2.2.1.2 Truck Graphs
- 2.2.2 TUM Trucks
- 2.2.2.1 Chrono Report
- 2.2.2.2 Truck Graphs
- 2.2.2.3 Calendar
- 2.2.1 Instantaneous Trucks
- 2.3 Why do my truck numbers look so high/low? What can I do about it?
- 2.4 Why are my TUM trucks higher than my Maximum Instantaneous trucks?
- 2.5 Is this schedule achievable with our truck fleet?
“Instantaneous Trucks” vs “TUM Trucks”
Firstly, it’s important to understand the difference between an “instantaneous truck” and a “TUM truck”. An “instantaneous truck” is operating and productive at an instant in time. A “TUM truck” is the number of truck hours in a period divided by the number of hours per truck in the period, based on the TUM.
In reality, trucks do not respect the TUM in an instant in time - in any one instant, the truck is either available or unavailable, not a mixture of the two. There’s a probability of whether a truck will be operating at an instant in time in the future (the TUM is the average over an extended period of time). This is the most correct way to think about trucks on an instantaneous basis and for very short term scheduling. The TUM accounts for but does not explicitly contain the frequency and duration of downtime events (a cyclone vs a broken bucket tooth).
In version 21.2, APS no longer supports TUM trucks and instead uses an approximation of instantaneous trucks. The Instant Units is a calculated field equal to the number of units entered, multiplied by the PA, UPA, OE, UA, and PF. Conceptually, the Instant Units is the exact number of trucks expected to be available for use at any instant over the schedule.
Although the Instant Units is the mode of the number of trucks expected to be available, the number of trucks that will be available on any day is based on a binomial probability distribution. However, there’s no practical way of using this probability distribution in APS, so we are just using an approximation of the mode (not an integer).
Field | Calculation | Value |
---|---|---|
Period Duration | - | 12 |
Number of Units | - | 20 |
Planned Availability [PA] | - | 80% |
Unplanned Availability [UPA] | - | 90% |
Use of Availability [UA] | - | 90% |
Operating Efficiency [OE] | - | 100% |
Productivity Factor [PF] | - | 100% |
Direct Utilisation | PA * UPA * UA * OE * PF | 80% * 90% * 90% * 100% * 100% = 64.8 % |
Engine Factor | [OE] * [PF] | 100% * 100% = 100% |
Hours / Unit | Period Duration * Direct Utilisation | 12 * 64.8% = 7.776 hours |
Instantaneous Units | Number of Units * PA * UPA * UA * OE * PF | 12.96 |
Truck Assignments
At the beginning of each period, APS determines the exact number of trucks expected to be available for use at any instant over the period. With the number of available trucks known, the trucks must be assigned to the mining Agents during each Steady State. The decision of how to assign trucks has now moved into the APS optimisation engine, taking this manual task away from the user.
The primary task of the APS optimisation is to maximise the amount of cash generated through mining movements whilst considering the constraints of the schedule. Mining movements equate to a particular dollar value, and some movements are worth more than others. In most cases, high-value movements involve sending material to the crusher. To prioritise higher value movements, we need to ensure Agents generating them are operating at maximum capacity by giving them sufficient trucks.
This process has been manual and involved users adjusting rates on a period by period basis. In APS 21.2, assigning trucks is automatic and works by prioritising Agents who generate movements with high revenue. As a simple example, if the highest revenue is for crusher feed movements, trucks will be prioritised to the Agents who are mining ore. The remaining trucks will then get assigned to Agents generating the next highest revenue movements. This process continues until either all of the available truck units have been assigned or if every Agent is operating at its maximum capacity. In many cases, lower priority Agents may not receive enough trucks causing them to fall short of their maximum operating capacity. When an Agent is under trucked, the Agents Productive Rate will automatically be adjusted to match the number of trucks assigned.
In summary, trucks are assigned to Agents based on how much money their movements are generating. If users wish to have full control and set the priority based on Agent, individual cash flows filtered by the individual Agents can be created. This setup can be useful when there is a duration where a particular Agent mining waste needs to be prioritised. The cash flow for that particular Agent can be increased so that its movements result in high revenue, bumping it up the priority list.
Limiting Trucks
Despite the APS optimisation automatically assigning trucks to Agents, users can place some restrictions. These limits can either be placed on individual Agents or on Truck Assignment Rules. The option to enable this functionality is located in the Final Configuration Step in the Setup Tab.
Agents
Limiting an Agent will enforce a hard constraint on the number of physical truck units the Agent can use during a single Steady State. In most cases, limiting the trucks for an Agent will cause the Agents rate to be reduced. The rate will be automatically adjusted to match the number of trucks specified—the image below highlights where this value can be set within the Calendar.
Truck Rules
Similar to Agents, limiting trucks within a Truck Assignment group enforces a hard constraint on the number of physical trucks available to be used within the group. Limiting trucks here can be useful for splitting trucks of the same model into different fleet groups.
Reporting Truck Usage
Both TUM trucks and Instantaneous trucks can be reported in APS; however, how each is reported is different.
Instantaneous Trucks
Reporting instantaneous trucks in traditional period-based reports simply doesn’t work due to the instantaneous time paradigm; however, there are several other places where the instantaneous trucks can be viewed and reported.
Steady States
Within the Steady States tab, trucking requirements for an Agent can be viewed at any instant in time. The field “Haulage.Instanatenous Trucks” reports how many trucks are currently assigned to the selected Agent.
Truck Graphs
In the Final Configuration Step, there is an Instantaneous graph option. When enabled, users will see the number of trucks each Agent was using and how many trucks are being used overall.
TUM Trucks
Reporting TUM trucks involves looking at consumed truck hours over the period. Users can find information on calculating truck hours in the PS. Truck Hour Calculation section.
Chrono Report
TUM trucks can be reported in a Chrono report by dividing the number of hours available per truck by the productive hours of that particular truck. The number of hours available per truck is available in the Calendar Data Source field list. Users can also find the total productive hours of a particular truck model Mining Data Source field list. A filter must be applied to return the productive hours of the required truck model. These fields and the calculation of truck hours is shown in the image below.
“Why can’t I report the number of trucks in a Pivot Report?”
You might think that you could divide the truck hours in each movement by the number of truck hours per unit in the schedule and sum that up in a pivot table. And you’d be right - but only if the sums are broken down into the scheduling periods as specified in the calendar - if not, it doesn’t work.
Truck Graphs
In the Final Configuration Step, there is a TUM truck graph option. When enabled, users will see the number of TUM trucks used over the period for each Agent and how many TUM trucks were used overall for the period.
Calendar
In the Calendar, under each truck model, both the productive hours consumed and the amount of TUM trucks this equates to can be seen. For a selected Agent, it is also possible to view how many TUM trucks of each truck model the Agent consumed.
The Fleet Use of Availability value represents the calculated Use of Availability of the fleet of trucks. This value is calculated without considering the [UA] input. The below example shows one way in which this value is calculated.
Hours per Unit = Period Duration * [PA] * [UPA] * [OE] * [PF]
Number of Trucks = Productive Hours / Hours per Unit
Fleet Use of Availability = Number of Trucks / Number of units
Why do my truck numbers look so high/low? What can I do about it?
Examine the animation, which displays the routes taken to ensure the correct paths are being used. If not, use Haul Infinity to ensure the correct routes are used. Take the time to ensure your haulage model is correct before starting to schedule.
Use a pivot report to ensure that the right destinations are used for each material type. If not, alter the destination rules.
If the numbers are too high, look to mine some material with shorter hauls.
If the numbers are too low, there’s an opportunity to mine some material with longer hauls.
Your last option, if you’ve tried the above, is to derate or shut down diggers. Often this cannot be avoided to meet short term ore requirements.
Why are my TUM trucks higher than my Maximum Instantaneous trucks?
If TUM availabilities are being used, this infers there are durations during the period where the truck is not being used for its primary task. Consider a closed-circuit which requires 3 trucks in which the truck type has a PA of 90%, UPA of 85% and UoA of 80%. To ensure there is always 3 trucks available on the run, the number of TUM trucks required is 4.3.
Field | Value |
---|---|
Period Duration | 12 hours |
Availability | 90% |
Unplanned Availability | 85% |
Use of Availability | 80% |
Hours per Unit | 12 * 0.9 * 0.85 * 0.80 = 7.344 |
Total Productive Hours | 32 |
Instantaneous Trucks (constant over period) | 2.686 |
TUM Trucks | 32 / 7.344 = 4.3 OR 2.686 / 0.9 / 0.85 / 0.8 = 4.3 |
Is this schedule achievable with our truck fleet?
Note this is an opinion. If this opinion is abhorrent to you or any of your stakeholders, refer to the previous question.
To determine whether the schedule is achievable with the truck fleet on-site, consider the Fleet Use Of Availability (UA), which is reported for each truck in the Calendar. Although many sites have a historical UA, this should not be an input (and is not when Dispatch mode is enabled). There is nothing wrong with varying the UA for the truck fleet each day, provided the objectives of the schedule are met. The reported fleet use of availability indicates how close to the edge the schedule is running. If the UA is at 100% fleet use of availability, then if a truck goes down, the schedule may not be achievable (it still may be if there’s an opportunity to catch up later). If it’s at 50%, half the expected trucks could be inoperable, and the schedule is still achievable.
Note that if UA for your site also includes times like when a truck driver does not show up for work, that should be instead be accounted for in either unplanned availability or one of the other availability fields. APS assumes that UA is historically the percentage of time that the truck could be used but intentionally wasn’t.