toolathon123/oceania-weekly-gov-efficiency

Oceania Weekly Government Service Efficiency

Weekly dataset of simulated public-service request records from an Australian government public-service agency (Canberra HQ, ~180 staff). This dataset is published on the Hugging Face Hub as the central, versioned source for weekly efficiency analysis, replacing ad-hoc Excel/database reporting.

• Region coverage: Australia (Oceania)
• Cadence: Weekly (ISO week 32, 2026)
• Language: English
• License: MIT (synthetic/mock data — no real personal information)
• Owner org size: medium public-service agency (50–200 staff)

Dataset Summary

Each row is one service request received through an agency service channel during the reporting week. Fields:

Column	Description
request_id	Unique request identifier (e.g. REQ-2026-0001)
region	Standard region code (ISO 3166-2:AU, e.g. AU-NSW)
service_type	Type of public service requested
request_date	Date the request was received (YYYY-MM-DD)
response_time_hours	Time to respond / handle the request, in hours (>= 0)
resolution_status	Status: resolved, closed, in_progress, pending, escalated, rejected
channel	Service channel: online, phone, email, in-person, mobile_app
staff_id	Anonymised staff identifier (STAFF-001…STAFF-180)
week_number	ISO week number of the reporting week (32)

Data Processing / Cleaning

Raw collection (800 records) is cleaned before publication:

1. Dropped invalid response times — records with empty or negative response_time_hours were removed (61 rows: 37 empty + 24 negative).
2. Standardised region — mapped state/territory names to standard Oceania Australia region codes (ISO 3166-2:AU): NSW→AU-NSW, Victoria→AU-VIC, Queensland→AU-QLD, Western Australia→AU-WA, South Australia→AU-SA, Tasmania→AU-TAS, ACT→AU-ACT, Northern Territory→AU-NT.
3. Added week_number — ISO week number of request_date (32).

Cleaned result: 739 records.

Weekly Efficiency Statistics — Week 32, 2026

Metric	Value
Total records (cleaned)	739
Average response time (hours)	25.09
Resolved / closed requests	706
On-time completion rate	95.53%

On-time completion rate = share of requests with resolution_status in (resolved, closed).

Resolution status breakdown

status	count
resolved	494
closed	212
in_progress	18
pending	11
escalated	3
rejected	1

Records by region

region	count
AU-NSW	252
AU-VIC	182
AU-QLD	131
AU-WA	87
AU-SA	48
AU-ACT	17
AU-TAS	12
AU-NT	10

Usage

from datasets import load_dataset

ds = load_dataset("toolathon123/oceania-weekly-gov-efficiency")
df = ds["train"].to_pandas()

avg_response_time = df["response_time_hours"].mean()
completion_rate = df["resolution_status"].isin(["resolved", "closed"]).mean()

Limitations

• Data is simulated/synthetic for demonstration; do not use for real operational decisions.
• Response times are aggregated at request level; SLA targets are not encoded in this version.