Back to main page

 

VAS Tool

(Vehicle Assignment and Scheduling Tool)

 

The VAS Tool is a program designed to create initial allocations of vehicles to farms. The tool then times the schedules that will result form the assignments. The tool provides two methods of improving the schedules created. These are a gradient descent improvement search and a Simulated Annealing improvement search.


Contents

Entering processing schedules

Changing inputs

Generating schedules

Timing of schedules

Van's schedule

Stock of modules

Performance measures

Improvement search


Top

Entering Processing Schedules

The VAS Tool creates initial truck timed schedules for a given processing schedule. The 'Import Schedule' button (shown in figure 1) is used to import a import a schedule from another file. The processing schedule can be entered manually via the cells coloured grey.  The information required from the processing schedule by the VAS Tool is given in figure 1.

 

1. Information required from the processing schedule.

Top

Changing Inputs

The user can select the vehicles available and their characteristics by changing the cells coloured grey in figure 2.

 

2. Specifying the available vehicles.

 

Other information that can be changed is highlighted by the grey cells in figure 3.

 

3. Inputs representing the system dynamics.

Top

Generating schedules

To generate a catching schedule for a given processing schedule the ‘Event Assignment’ button (figure 4) should be clicked. This prompts the ‘Number of random assignments performed’ window (figure 5) to appear. From experience, good schedules were found with 100 random assignments.

 

4. ‘Event Assignment’ button.

 

5. ‘Number of random assignments performed’ window.

Top

Timing of Schedules

A sheet is produced for each available truck. These sheets display the events each truck will undergo. An example is shown in figure 6.

Farm

Demand

Trip number

Modules delivered/ picked up

Catching gang start

Catching gang depart

Farm arrival

Start catching

Farm departure

Plant arrival

Plant start unloading

Plant departure

Farm 1

21

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Farm 2

11

1

11

3

3

2:13 a.m.

2:20 a.m.

3:24 a.m.

3:38 a.m.

3:38 a.m.

4:17 a.m.

 

 

 

 

 

 

 

 

 

 

 

 

Farm 3

22

1

22

3

3

4:27 a.m.

4:34 a.m.

7:14 a.m.

7:24 a.m.

7:24 a.m.

8:13 a.m.

 

 

 

 

 

 

 

 

 

 

 

 

Farm 4

35

1

22

8

8

8:25 a.m.

9:27 a.m.

10:15 a.m.

10:27 a.m.

10:27 a.m.

10:56 a.m.

6. The timing of truck 1’s schedule

Top

Van's Schedule

There is a sheet that records when the van should arrive each farm and when it should depart each farm.

Stock of Modules

There is a sheet that records the stock of modules at the plant. The stock of modules is recorded each time a truck starts unloading at the plant and each time a truck departs the plant. This enables the plant idle time to be calculated. The plant idle time occurs when the plant runs out of chickens to process (full modules).

Top

Performance Measures

There is a sheet that summarises the schedule. Included in the sheet are statistics on the schedules performance. These performance measures are displayed in figure 7. The first row of the results are in minutes. The last plant arrival is given as 12 hour am/pm time in the bottom row. The bottom right hand corner of the table displays the 'combined total shift time' in hours and minutes.

7. Performance measures of the schedules.

 

These performance measures can also be found for schedules entered by hand rather than computer generated schedules.

Top

Improvement Search

The VAS tool provides both a gradient descent improvement search option and a Simulated Annealing improvement search option. The user enters how many iterations of the gradient descent should be performed. Once the iterations have been formed the best performing schedule found is entered into a sheet.

The Simulated Annealing improvement search option has the interface displayed in figure 8.

 

8. Simulated Annealing interface.

 

The initial temperature chosen will depend on how much the user wants the search to be diversified. The number of iterations performed at each temperature effects how much of the solution space is searched. While the decrement factor effects the intensification of the search. That is, the lower the decrement factor the faster the search will become intensified. The time entered into the box controls the length of the search. The Algorithm will continue searching until the time is up.

    The cooling schedule used in Figure 8 was found to be a good combination to use for the VAS Tool. Once the improvement phase has been competed, the best schedule found is stored in a sheet.

 

Back to main page