Datasets:
operating_conditions dict | question stringclasses 3
values | calculation_assumptions stringclasses 3
values | reference_answer dict | allowed_error dict | accepted_answer_range dict |
|---|---|---|---|---|---|
{
"equipment_type": "Hydraulic excavator",
"equipment_model": "Mid-size excavator with a 1.2 m³ general-purpose bucket",
"bucket_capacity": 1.2,
"bucket_capacity_unit": "m³",
"cycle_time_seconds": 24,
"bucket_fill_factor": 0.9,
"job_efficiency": 0.75,
"material_type": "Moist granular soil",
"swell_fac... | What is the expected bank-volume production in m³ per scheduled hour? | Calculate loose production as bucket capacity × fill factor × (3600 ÷ cycle time) × job efficiency. Convert loose volume to bank volume by dividing by the swell factor. Job efficiency is applied once, and the result is per scheduled hour. | {
"value": 97.2,
"unit": "m³ bank/h"
} | {
"absolute_tolerance": 1,
"relative_tolerance": 0.01,
"tolerance_mode": "both"
} | {
"lower_bound": 96.228,
"upper_bound": 98.172,
"unit": "m³ bank/h"
} |
{
"equipment_type": "Wheel loader",
"equipment_model": "Wheel loader with a 2.4 m³ general-purpose bucket",
"bucket_capacity": 2.4,
"bucket_capacity_unit": "m³",
"cycle_time_seconds": 42,
"bucket_fill_factor": 0.85,
"job_efficiency": 0.8,
"material_type": "Well-graded crushed aggregate",
"swell_factor... | What loose-volume production rate should be expected in m³ per scheduled hour? | Calculate production as bucket capacity × fill factor × (3600 ÷ cycle time) × job efficiency. The cycle time already includes the stated short-haul travel, so no additional travel-time adjustment is applied. Report loose volume; do not convert using the swell factor. | {
"value": 139.8857142857143,
"unit": "m³ loose/h"
} | {
"absolute_tolerance": 1.5,
"relative_tolerance": 0.01,
"tolerance_mode": "either"
} | {
"lower_bound": 138.3857142857143,
"upper_bound": 141.3857142857143,
"unit": "m³ loose/h"
} |
{
"equipment_type": "Hydraulic excavator",
"equipment_model": "Compact excavator with a 0.8 m³ trenching bucket",
"bucket_capacity": 0.8,
"bucket_capacity_unit": "m³",
"cycle_time_seconds": 18,
"bucket_fill_factor": 0.95,
"job_efficiency": 0.7,
"material_type": "Sandy clay",
"swell_factor": 1.2,
"ha... | How many scheduled operating hours are required to excavate 620 m³ bank? | Calculate loose production as bucket capacity × fill factor × (3600 ÷ cycle time) × job efficiency, then convert to bank production by dividing by the swell factor. Scheduled hours equal the required bank volume divided by bank production per scheduled hour. Do not apply job efficiency a second time. | {
"value": 6.992481203007519,
"unit": "scheduled h"
} | {
"absolute_tolerance": 0.1,
"relative_tolerance": 0.02,
"tolerance_mode": "absolute"
} | {
"lower_bound": 6.892481203007519,
"upper_bound": 7.092481203007519,
"unit": "scheduled h"
} |
Earthmoving Productivity Calculation Answer Verification Benchmark
This benchmark focuses on earthmoving equipment capacity, productivity, cycle operations, and work-quantity calculations. Each record provides structured operating conditions, a calculation question, assumptions, a reference numeric answer, and an allowed error range. The JSON format supports parsing problem conditions, checking results and units, and automatically determining whether an answer meets the specified tolerance. It is suited to evaluating construction equipment calculations, analyzing earthmoving work plans, and training or validating related models; sample count and licensing terms should be confirmed in the delivery documentation.
Introduction
Evaluate earthmoving calculation skills with structured operating conditions, gold answers, and explicit tolerance rules.
Dataset details
- Sample rows in this repo: 3
- Full dataset size: 6023 piece
- Full dataset file size: 0.4 GB
- Modality: text
- Language: en
- Industry: Construction & Earthmoving Machinery
Upload this folder to a Hugging Face dataset repository. samples.jsonl is the sample split. The full dataset size above is the listing size, not the number of rows in the file.
Fields
| Field | Description |
|---|---|
operating_conditions |
{'title': 'Operating Conditions', 'description': 'Structured equipment, material, and job parameters used to solve the earthmoving productivity problem.'} |
question |
{'title': 'Calculation Question', 'description': 'Question asking for an earthmoving productivity, production, cycle, or work-quantity result.'} |
calculation_assumptions |
{'title': 'Calculation Assumptions', 'description': 'Assumptions, conventions, and calculation conditions used to interpret the problem and its units.'} |
reference_answer |
{'title': 'Reference Answer', 'description': 'Gold-standard numeric result and its unit for the calculation question.'} |
allowed_error |
{'title': 'Allowed Error', 'description': 'Error limits and comparison rule used to determine whether a submitted numeric answer is acceptable.'} |
accepted_answer_range |
{'title': 'Accepted Answer Range', 'description': 'Derived numeric interval and unit representing the permitted answer range under the configured error criteria.'} |
FAQ
What does the benchmark contain?
Each record includes structured equipment, material, and job conditions; a calculation question; calculation assumptions; a reference numeric answer; allowed-error criteria; and the resulting accepted answer range. The fields are organized as JSON for programmatic access and evaluation.
How are submitted answers evaluated?
Parse the submitted result according to the units and calculation conditions specified by the problem, then compare it with the reference answer using the record's allowed-error criteria and comparison rule. The accepted answer range can help determine whether a numeric result meets the threshold; evaluation systems should still apply the specific error rules when handling unit conversions and number formats.
What are suitable use cases?
The benchmark can support automated evaluation of earthmoving equipment productivity and work-quantity calculations, validation of numerical reasoning, engineering equipment education, and regression testing for construction parameter analysis tools. Reference answers follow the assumptions in each record and are not a substitute for field measurements or professional engineering review.
What are the dataset size and licensing terms?
The available product information does not specify the sample count, file size, or licensing terms. Confirm these details against the actual delivery manifest and license agreement before purchase or integration, including permissions for training, evaluation, and commercial use.
For more dataset information, please visit https://www.mobiusi.com or email contact@mobiusi.com
- Downloads last month
- 73