Rendering cover…
Free
Rendering cover…
Free access
DT / SR-06 / 2026-09
Electric Three-Wheelers & Motorbikes — Free Dataset | DT Linux
- Author
- K. H. Militha Mihiranga
- Organisation
- Data Tune (DT Linux)
- Issued
- 19 September 2026
- Version
- 1.0
- Pages
- 35
- Licence
- Free to download and reuse with attribution
- Downloads
- 0
Data provenance
Free to download and reuse with attribution to Data Tune (DT Linux). Not advice to any individual vehicle owner — see the constraints on use.
Outputs
Report / paper (PDF)
Data-Tune-Insight-Report-Electric-Motorbikes-and-Three-Wheelers
Twenty sections covering collection architecture, the discounted cost model, three layers of validation, the crossover result by segment, parity across the fleet, diffusion fitted to sixty months, the charging coverage gap, the grid peak problem and three conditional scenarios to 2035. Eighteen figures.
Dataset (XLSX / CSV)
Data-Tune-SR-06-EV-Transition-Dataset
The full registration panel, price series, crossover table, parity shares, sensitivity sweeps, backtest scores, charging coverage, hourly grid profiles and scenario paths. 1,822 formulas, all recalculating.
Pipeline scripts / code
Data-Tune-SR-06-Reproducible-Pipeline
analysis.py, charts.py, build.py and workbook.py with the source panel, price series, computed statistics and the intake record. Includes the runner, so the whole 35-page report rebuilds with one command in about thirteen seconds.
Pipeline scripts / code
Data-Tune-SR-06-Figures-SVG
Every chart as scalable vector graphics, named to its figure number, including the crossover money chart and the tornado sensitivity analysis.
At a glance
The study in six numbers
Every one of these is reproducible from the delivered dataset.
5,332,229registrations across 4 markets and 60 months
7,378–9,520km a year — where electric becomes cheaper
100–100%of commercial fleets are already past that line
52–54%of private vehicles — the same machines, half the case
45%charging coverage weighted by commercial operating density
157 MWadded at system peak by 2035, against 11 MW if managed
Findings
What the research found
Five findings, each traceable to a section of the report.
01
Utilisation decides the transition, not purchase price
An electric two- or three-wheeler converts a capital premium into a running saving, so the trade is only as good as the number of kilometres it is divided by. Crossover mileages vary by only 1.3× across five quite different vehicles — but median annual mileage varies by 4.1×. All the variation that decides outcomes is on the utilisation side.
02
Commercial three-wheelers are finished as an economic question
Hire, delivery and goods fleets run at a cost advantage of 29–35% and clear the threshold by 2.6× or more. Private motorcycles and private three-wheelers sit at 0.4–2.0% — within a rounding error of break-even.
03
Fuel price alone can flip the sign
A pump-price swing from $0.70 to $1.45 moves the electric advantage by 3.64 cents per kilometre against a base advantage of 3.18 cents — 114% of base, more than every other assumption combined. Any crossover figure quoted without its pump price is an opinion about oil markets wearing an engineering result's clothes.
04
Once the economics are settled, they stop being the constraint
For the three commercial segments the cost question is closed, and adoption still has not completed. Charging coverage weighted by operating density is 44.7%, with only 5 of 60 districts meeting a fifteen-minute-detour requirement — and the shortfall is worst exactly where the ready fleet works.
05
The cheapest intervention is about timing, not quantity
The system peak falls at 18:00, which is also when drivers plug in on getting home. By 2035 the fleet adds 157 MW at peak if charging is unmanaged and 11 MW if it is shifted overnight. Same energy, same journeys, 14× the capacity.
Results
Where the line falls, by segment
Discounted whole-life cost per kilometre, Sri Lanka prices at $1.05 per litre.
| Segment | Use | Crossover (km/yr) | Median (km/yr) | Headroom | Cost advantage | Fleet past parity |
|---|---|---|---|---|---|---|
| Three-wheeler - passenger hire | commercial | 9,520 | 31,800 | 3.34× | +35.0% | 100% |
| Motorcycle - delivery / courier | commercial | 8,002 | 27,500 | 3.44× | +35.1% | 100% |
| Three-wheeler - goods | commercial | 8,928 | 23,400 | 2.62× | +29.2% | 100% |
| Motorcycle - private | private | 7,378 | 7,800 | 1.06× | +2.0% | 54% |
| Three-wheeler - private | private | 9,105 | 9,200 | 1.01× | +0.4% | 52% |
Headroom is median mileage divided by crossover mileage — above 1.00 the median vehicle is past the economic threshold. The three commercial segments clear it by 2.6× or more; the two private segments sit at 1.06× and 1.01×.
Methodology
How the research was done
The same four-step method Data Tune applies to every data collection and data mining engagement.
Step 01
Collection
Registrations by market, segment and powertrain over 60 months, plus matched retail prices, fuel and tariff series, battery pack prices, a charging estate snapshot and a fleet utilisation survey of 10,800 vehicles.
Step 02
Cost modelling
Discounted total cost of ownership per kilometre over the full vehicle life, including battery replacement placed by cumulative distance, residual value and end-of-life handling. Solved for the crossover mileage.
Step 03
Mining
Crossover applied against the observed mileage distribution to give parity shares; Bass diffusion fitted to each country-segment series; charging coverage measured against commercial operating density.
Step 04
Validation
Three layers: a 12-month backtest against naive and drift baselines, a four-way assumption sweep, and a one-at-a-time tornado showing which assumption the conclusion actually rests on.
Constraints on use
- This is not advice to any individual vehicle owner. Three-wheeler driving is low-income self-employment across all four markets studied. A fleet-level cost finding says nothing about whether a particular driver — facing a particular credit offer, route and household balance sheet — should switch.
- No manufacturer, model or brand is named, compared or recommended anywhere in this research. Vehicles are described by segment and duty cycle only.
- The research takes no policy position. Duties, subsidies and tariffs on these vehicles are politically contested in every market here. The study presents the arithmetic under stated assumptions; choosing between the options it informs is not its role.
- Battery end-of-life is a hazard, not a line item. Informal lead and lithium recycling causes documented harm. It is costed here as a TCO term, which is an accounting treatment rather than a solution.
- The scenarios are conditional, not forecasts. Each states what would falsify it, and none may be quoted as a prediction of 2035.
Questions
Frequently asked questions
About the data, the assumptions and what the result does and does not show.
Is this dataset really free to download?
Yes. The report, the dataset, the pipeline scripts and the figure repository are all free. There is no account to create, no payment and no email form. Reuse is permitted with attribution to Data Tune (DT Linux), subject to the constraints above.
Aren't cheap electric motorcycles supposed to lead this transition?
That is the common assumption and this study finds the opposite. Private motorcycles are the cheapest vehicles and the lowest-mileage ones, so they sit at 1.06× headroom — essentially at break-even. Commercial three-wheelers cost far more and cover 4.1 times the distance, which puts them 3.3× clear. Volume and economic readiness are in different segments.
How sensitive is this to your assumptions?
Very, and the report publishes the sweep rather than hiding it. Fuel price alone moves the electric advantage by 114% of its base value — more than battery life, pack price, discount rate and residual value combined. The commercial segments stay past parity across every swept range; the private segments cross back and forth within it.
Does the adoption model actually predict anything?
Modestly, and it was tested. Fitted on 48 months and scored on 12 held out, it achieved 1.97 percentage points of error against 2.52 for assuming no change — +17.1% skill, beating the baseline in 15 of 20 series. That justifies extrapolating the curve's shape. It does not justify predicting 2035, which is why that section is written as conditional scenarios.
Can Data Tune run this on our market or our fleet?
Yes. The pipeline can be re-pointed at licensed registration and price data for your market, or at your own vehicles, mileages, financing rate and tariff — producing a ranked conversion order rather than a segment average. Email info@dtlinux.com or call +94 77 527 1186.
Capability
Services behind this research
Data Tune builds custom datasets, mines them and delivers the analysis. Research outsourcing for teams without an in-house data function.
Data CollectionAPI, log, field and web sources, structured to your schema
Data MiningCustom scraping, directory extraction, lead generation data
ML DatasetsLabelled, validated training data for AI models
IoT SolutionsSensor pipelines and live monitoring dashboards
MarTech SolutionsMarketing data engineering and attribution
Data StoreReady-made datasets available for purchase
Want the crossover for your own fleet?
Send us your vehicles, mileages, financing rate and tariff and we will scope a live study on the same method — producing a ranked conversion order, a charging shortfall list and a peak-load profile you can take to your utility.
Consultant
K. H. Militha Mihiranga
Data Engineer · Data Solutions Consultant
Office
555/24 Ranmuthugala,
Kadawatha, Sri Lanka
Office hours
Kadawatha, Sri Lanka
Monday to Friday, 9 AM – 5 PM (UTC+5:30)
Email
Telephone
Website
Contact
K. H. Militha Mihiranga
Data Engineer · Data Solutions Consultant
555/24 Ranmuthugala,
Kadawatha,
Sri Lanka
Kadawatha,
Sri Lanka
More research
DT / SR-02 / 2026-09
Cross-Platform Social Sentiment Tracking — Free Dataset & Report | DT Linux
PDF
XLSX
HTML
View & download
8 downloads
Rendering cover…
Free