Case Study: Aspendale Level Crossings — 4 Visits in 5 Days

Summary

Carrum Signals Section 4201 visits Aspendale 4 separate times in 5 days (19-23 June 2026) to perform work that could be consolidated into a single visit. This example illustrates how the current scheduling approach generates unnecessary travel by treating each work order independently.

The Schedule (As Planned)

The June 21 visit is part of a larger 10-WO day covering Aspendale (2), Edithvale (4) and Kananook (4). The June 22 visit is the team's entire day — a single auto gate inspection at one level crossing.

Why This Happens

All five work orders at Aspendale are for level crossing protection equipment (auto gates and escape gate latches) with a 61-day frequency and 6-day tolerance. The scheduling system calculates each order's window independently and places the planned date at the midpoint. Because each crossing's maintenance cycle is offset by a day or two, the midpoints land on different days:

The common overlap window across all five orders is 17-28 June — 11 days in which all work could be completed on a single visit.

Single Line, Walking Distance

Aspendale sits on the Frankston line — a single, straight-through rail corridor with no junctions or branches. The level crossings in question are road crossings over this single line, spread across just 1.7km of track. The two crossings inspected on June 21 are only 364 metres apart.

These are trackside inspections (SP387 — "Inspect, Check & Test" of auto gates) that do not require a track shutdown or possession. There is no operational reason — different lines, different access requirements, or different shutdown windows — that would prevent all crossings from being inspected on the same visit. A technician could walk between them in minutes.

The Inefficiency

• 4 separate trips from Carrum depot to Aspendale (~7km each way)
• The June 22 visit sends a team for a single work order — one auto gate inspection at one level crossing, which is the team's entire workload for the day
• The same job type (SP387) is performed on 21 Jun AND 22 Jun, at different crossings just hundreds of metres apart on the same rail line
• All tolerance windows overlap, so there is no technical reason these must be on different days
• The crossings do not require a shutdown — these are routine trackside inspections

What an Optimised Schedule Would Look Like

All 5 Aspendale work orders could be batched into the June 21 visit, which already has the team on-site doing the same type of work. Combined with the existing Edithvale and Kananook work on that day, this would bring the June 21 total to 15 WOs — above average but within normal range (teams regularly complete 12+ WOs in a day).

This would eliminate 3 unnecessary visits, saving approximately 42km of travel (3 × 7km each way) and the associated access/protection arrangements for each visit.

The Broader Pattern

This pattern repeats at Aspendale later in the year:

The same team visits Aspendale 6 times in 12 days in August. The same scheduling behaviour — each work order placed independently at its window midpoint — creates the same fragmented pattern.

Across the full year, Carrum 4201 visits Aspendale on 27 separate days. With tolerance window optimisation, this could be reduced to approximately 9 visits — a 67% reduction.

Relevance

This case study demonstrates that:

1. The scheduling system applies no geographic batching — it does not consider what other work exists at the same location when placing a work order
2. The tolerance windows are wide enough to consolidate — even with a tight 6-day tolerance on 61-day frequency jobs, there is sufficient overlap
3. The same team performs all the work — there is no competency barrier to consolidation. Carrum 4201 does SP387, SP395, SP206, SP305A, SP311, SP375, and SP208 throughout the year
4. Single work order visits are common — 31% of all location visits across the network are for just 1 work order, and 96% of these could be merged with another visit
5. The inefficiency compounds — Aspendale is one of 268 group+location combinations where visits could be consolidated. Across the network, 73% of location visits could be eliminated through tolerance window optimisation