Truck maintenance ERP for a 25-truck Surat fleet: job cards, tyres, batteries and AI truck health, a case study
Looking for truck maintenance software or a fleet maintenance ERP in Surat or Gujarat? Yukti AI, a software company in Varachha, Surat, builds custom ERP, automation and AI for transporters and truck-fleet owners. This page shows a real build: the truck maintenance ERP made for Rajeev Roadlines, a Surat fleet of 25 trucks running external freight, group-company haulage, contract rental and material trips. It treats each truck as an asset and answers one question for the owner: what does this truck cost to keep on the road, and why? Job cards with ten stages, service schedules by kilometre and by days, breakdowns with downtime, tyres tracked by axle position through retreading, godown and scrap, batteries with health and warranty, spare parts that deduct from stock, vendor and mechanic comparison, bills, payments and driver advances, 25 reports and a rule-based AI layer that scores every truck’s health and shows its evidence. GPS, FASTag, fuel cards, WhatsApp and Tally are not connected in this version, and no results are claimed. A custom ERP from Yukti AI starts from ₹30,000; the final quote for a fleet of this breadth comes after a free 30-minute workflow audit.
This is the working version of a truck maintenance ERP built around a 25-truck Surat fleet’s workflow, with records kept in the browser. It covers trucks with a 12-tab profile, drivers, trips and trip costing, trip and repair expenses with approvals, job cards, service schedules, breakdowns, tyres, batteries, spare parts, scrap, payments and bills, vendors and mechanics, contracts, documents, reports, notifications, a driver app and four roles. The AI layer is rule-based intelligence on the fleet’s own records: it finds the costly truck, the overdue service, the tyre with four punctures in a month and the garage whose repairs come back, explains each finding with evidence and a rupee estimate, scores every truck from 0 to 100, answers questions in English and transliterated Gujarati, and prints a monthly report. There is no language model in it today; one can be added behind the same interface. Live GPS, FASTag, fuel cards, WhatsApp alerts, Tally and GST billing are not connected and are listed on this page as additions, not as delivered.
Truck maintenance ERP software in Surat – enquire
Yukti AI, a software company in Surat, builds truck maintenance ERP, fleet management software, WhatsApp automation and AI for transporters in Surat, across Gujarat and across India. The case study below shows a real build; this box tells you what you get and how to start.
- What you get: job cards, service schedules, breakdowns, tyres, batteries, spare parts, vendor bills and payments, driver advances, trips and trip costing, documents with expiry alerts, 25 reports, a driver app and a rule-based AI layer that scores truck health, with WhatsApp, GPS, FASTag and Tally added when you are ready.
- Built around your fleet: your axle configurations, service intervals, expense categories, garages, roles and approval rules, not a generic template.
- From price: a multi-module ERP starts from ₹30,000; the final quote for a fleet of this breadth is given after a free 30-minute workflow audit.
- Where we are: 220, Leonard Square, Yogichowk, Varachha, Surat – 395006, Gujarat.
- Next step: send us one month of garage bills and your service diary for two trucks, and we will show what one system would replace.
Contents
- Quick answer and quick facts
- Screenshots
- The business: where a fleet’s repair money goes and who this fleet maintenance ERP is for
- The challenge: why diaries, garage bills and WhatsApp stop being enough
- How a fleet’s maintenance work changes, task by task
- Trucks, the 12-tab truck profile and drivers
- Job card software for truck repair: the 10-stage workflow
- Service schedule by km and days and breakdowns and downtime
- Tyre management across axle positions, retreading, godown and scrap and battery tracking
- Spare parts stock and scrap and vendor and mechanic comparison
- Expenses, approvals, bills, payments and driver advances
- Trips, mining trip costing and contracts and documents and expiry
- Dashboard, 25 reports and notifications
- The driver app and roles
- Worked example: maintenance cost per km, tyre cost per km and the truck health score
- AI for a truck fleet: rule-based intelligence that shows its evidence and what a truck health score is
- AI and automation: built here, built in other Yukti systems, buildable, not built and what WhatsApp, GPS, FASTag and Tally would add
- Custom fleet ERP vs GPS tracking apps vs ready-made fleet software vs Excel
- Features checklist, benefits and buying checklist
- Truck maintenance software for Surat and Gujarat fleets
- Solution architecture, the modules and how it was built
- What was delivered and what this page does not claim
- Technology, cost and timeline and which Yukti fleet build fits you
- FAQ
What is truck maintenance ERP software, and what should it include for a 25-truck fleet?
Truck maintenance ERP software is a system that records everything spent on keeping each truck on the road — job cards, services, breakdowns, tyres, batteries, spare parts, garage and vendor bills — against that truck and its kilometres, so the owner can see the cost per km of every vehicle, what is due, what is overdue and which garage or tyre brand is costing more than it should. It is different from GPS tracking, which tells you where a truck is, and from freight billing software, which tells you what a trip earned. For a 25-truck fleet it should include a truck profile with every document and expiry, job cards with stages and parts, a service schedule by km and by days, breakdowns with downtime, tyre and battery lifecycles, spare parts stock, vendor bills and payments, driver expense approval, reports by truck and a driver app. Yukti AI, a software company in Surat, built exactly this for Rajeev Roadlines: 22 modules, a 12-tab truck profile, 10-stage job cards, 13 service types, 12 tyre event types, 25 reports, four roles and a rule-based AI layer with a truck health score. A custom ERP from Yukti AI starts from ₹30,000, and a first working version is typically ready in about 7–10 days.
Can fleet software predict breakdowns without GPS or sensors?
Partly, and honestly: from records, not from the engine. Without sensors a system cannot read engine temperature or tyre pressure, but it can read what the fleet already writes down: kilometres per trip, services done and skipped, repeat repairs of the same category, punctures per tyre, breakdowns per truck, battery health readings and the age and usage of each vehicle. The Rajeev Roadlines build turns those records into seven kinds of predictions — oil change, service, tyre replacement, battery, part failure, breakdown risk and document expiry — each with its reason, history, due date or km, estimated cost and a confidence value, and each labelled as an estimate. That is a records-based early-warning system, not a digital twin. When a fleet adds GPS, OBD or tyre-pressure sensors later, the same predictions get better inputs.
Quick facts
| Client | Rajeev Roadlines, a truck fleet owner and transport contractor in Surat, Gujarat |
|---|---|
| Fleet the system was built for | 25 trucks (per the brief): 6-wheel, 10-wheel and 12-wheel configurations; owned, financed and leased vehicles; external freight, internal group-company haulage, contract rental and material trips such as sand and aggregate |
| Problem | Repair money spent across diaries, garage bills, tyre shops and WhatsApp with no per-truck cost, services done from memory, tyres and batteries with no life record, parts bought twice, garages judged on price alone, and expiries noticed too late (our description of the gaps, not the client’s words) |
| Solution | A custom truck maintenance ERP: trucks, drivers, trips, expenses, approvals, job cards, service schedule, breakdowns, tyres, batteries, spare parts, scrap, payments and bills, vendors and mechanics, contracts, documents, reports, notifications, settings, a driver app and an AI intelligence centre on one dataset |
| Core mechanics | One analytics layer feeds every screen; job-card parts deduct from stock; services recompute their next due km and date; bills flow to payments; tyre events move a tyre between positions, godown, retreading, sale and scrap; the health score and every AI insight link back to the record they came from |
| Screens | 22 sidebar modules, a 12-tab truck profile, a 9-tab AI intelligence centre, a driver app with 7 tiles and a global search |
| Counts | 10 job-card stages, 13 service types, 10 trip-expense and 19 repair categories, 12 tyre and 10 battery event types, 13 part categories, 7 stock transaction types, 11 vendor types, 14 document types, 4 axle configurations, 25 reports |
| Roles | 4: Super Admin / Owner, Admin / Accountant, Maintenance Manager, Driver |
| AI | Rule-based intelligence on the fleet’s own records: 20 insight families, anomalies, 7 prediction kinds, a 7-component truck health score, a route-adjusted driver index, chat in English and transliterated Gujarati, a monthly PDF; no language model connected |
| Stack | React 18, TypeScript 5, Vite 5, Tailwind CSS v4, Zustand, IndexedDB persistence, Recharts, three.js, jsPDF, SheetJS |
| Connections | GPS / telematics, FASTag, fuel cards, WhatsApp, Tally, e-way bill and GST billing, a server database and login are not connected; each is listed below as an addition |
| Status | The working version built by Yukti AI around the fleet’s workflow, records in the browser; usage and results are not claimed |
What the fleet’s office and maintenance team work in
Screens from the truck maintenance ERP built by Yukti AI for Rajeev Roadlines. Figures, truck numbers, driver, garage and vendor names visible in the screenshots are from the sample dataset the system ships with; they illustrate the screens and are not the company’s business numbers.








In a truck fleet, the repair money disappears between the diesel pump and the garage
A truck earns on the road and spends in the workshop, and most fleet owners can tell you the first number and only guess the second. Freight income arrives per trip and is written down because somebody has to be billed. Maintenance leaves the business in small pieces: a clutch plate at a roadside garage on the highway, two tyres from the tyre shop, a battery under warranty that was never claimed, a puncture paid by the driver from his advance, engine oil from the parts dealer, labour to the mechanic in cash, the denter, the welder, the electrician. Each piece has a bill, or a WhatsApp photo of a bill, or nothing. By the end of the month the owner knows the fleet spent a lot and cannot say which truck, which part, which garage or whether the same fault came back for the third time.
The fleet this system was built around is a Surat transporter with 25 trucks across 6-wheel, 10-wheel and 12-wheel configurations, some owned outright, some financed, some leased. The trucks run four kinds of work: external freight for customers, internal haulage for group companies, trucks on contract rental to a client by the month, per trip or per km, and material trips where the truck buys sand or aggregate at a mine or crusher with a mining challan and sells it on. Services are done at the yard and at outside workshops; tyres are bought new, retreaded, moved between trucks, kept in the godown and eventually sold as scrap; parts are bought from suppliers, taken from stock, reused from another truck or reconditioned. Drivers pay for diesel, toll, loading and punctures on the road from advances and submit the bills afterwards.
Yukti AI’s job was to put the whole maintenance side of that business — the truck as an asset — into one system that the owner, the accountant, the maintenance manager and the drivers all work in, and to add a layer of rule-based intelligence that reads those records and points at what needs attention. The trip and freight-billing side of a transport business is covered in our other fleet build, the Shree Momai fleet management ERP; this build is its complement.
A truck maintenance ERP has to attach every rupee to a truck and a kilometre: the job card, the service, the tyre, the battery, the part and the garage bill must all land on the vehicle that used them, so that cost per km, health and “which garage is really cheaper” are facts, not impressions. Billing the customer is a different system’s job.
Who is this fleet maintenance ERP for?
The system was designed around a Gujarat transporter with a fleet of about 25 trucks, its own yard, outside garages and a mix of freight, group haulage, contract and material work. If you own and run between 10 and 100 trucks — a transport contractor, a tipper and dumper fleet serving mines, crushers and construction sites, a trailer fleet on port and industrial runs, a logistics company with its own vehicles or a manufacturer with a captive fleet — the problems on this page will sound familiar. It is also a worked answer to the search for fleet management software that goes deeper than GPS dots on a map: software built around what a truck costs to keep running.
It is a good fit if most of these are true for you:
- You cannot say, for any truck, what it cost in maintenance per kilometre last month, or which of your trucks is the expensive one and why.
- Service due dates live in a diary, on a whiteboard in the yard or in the driver’s head, and the oil change happens when somebody remembers.
- Tyres are bought, swapped between trucks, retreaded and scrapped with no record of which tyre is where, how many kilometres it gave or which brand punctures most.
- Batteries are replaced before the warranty is used, because nobody knows when it ends.
- Spare parts are bought again because the godown stock is unknown, and parts used on a repair are never taken off the stock.
- The garage with the lowest estimate gets the work, and the same fault comes back within a month.
- Driver-submitted diesel, toll and puncture bills are approved from a WhatsApp photo, with no comparison across drivers on the same route.
- Insurance, PUC, fitness, permit, road tax and driver licence expiries are discovered at a checkpost or by a fine.
It is not for a fleet whose only need is live tracking (buy a GPS service first), for a transport broker with no owned vehicles, or for a single-truck owner-operator. If your bigger problem is trips, diesel evidence, document ladders, EMIs and GST invoicing, read the Momai fleet ERP case study first; the two builds are designed to complement each other, and a fleet can have both in one system. Yukti AI is in Surat and builds these systems for transporters across Surat, Gujarat and India; see AI and software for logistics and transport and custom ERP software in Surat.
Why diaries, garage bills and WhatsApp stop being enough for a 25-truck fleet
These points are our description of the gaps a system of this kind has to close, drawn from how truck fleets in Gujarat run their maintenance and from the scope of the build. They are not quotes from the client, and no measurement of the company’s previous way of working is claimed.
- Repair cost is never per truck. Garage bills, parts bills and cash to mechanics are recorded, if at all, as a monthly total. Nobody can rank the fleet by maintenance cost per km, so the truck that quietly eats the profit is invisible until it is sold.
- Services run on memory. An engine oil change is due by kilometres, a greasing by days, a filter by whichever comes first. With odometer readings in trip sheets and due dates in a diary, services slip, and a slipped service becomes a breakdown on NH-48 at night.
- Tyres are the second-largest running cost and have no biography. A tyre moves from a new purchase to a front axle, to a rear position after rotation, to the retreader, back to another truck, to the godown and to a scrap dealer. Without a record per tyre, tyre cost per km and brand comparison are impossible, and punctures are just an expense line.
- Batteries are replaced, not managed. Health, voltage and water level are checked when the truck fails to start. Warranty claims are missed because the purchase date and warranty months are on a bill in a file.
- Parts are bought twice. The godown holds filters, bulbs, belts and brake parts nobody counted. A job card uses parts that are never deducted, so stock on paper and stock on the shelf drift apart.
- Garages are judged on the estimate. The cheapest quote wins, and the fact that its repairs come back within thirty days is not written anywhere. The gap between the estimate and the final bill is remembered as a grievance, not as a number.
- Driver expenses have no fair comparison. Diesel per trip is compared driver to driver without accounting for the route, the load or the truck class, so the driver on the hilly or heavy route looks worse than he is, and the driver padding toll claims looks fine.
- Money is a separate world. Vendor bills, due dates, driver advances and receivables are in the accountant’s ledger or in Tally, which knows nothing about job cards, tyres or trucks.
How does a fleet’s maintenance work change, task by task?
The table walks through the real jobs in a truck fleet’s yard and office and what each looks like in the ERP. The “before” column is a general description of the manual way of working, not a measurement, and no time savings are claimed.
| Task | Before (manual) | After (in the system) |
|---|---|---|
| A driver reports a fault | Phone call or WhatsApp voice note to the owner | Breakdown or complaint from the driver app, against his truck, with location, highway and photos; the office opens a job card from it |
| Getting a repair done | Garage estimate on the phone, work done, bill later | Job card through ten stages: created, inspection, approved, in progress, parts required, completed, tested, delivered, bill received, paid; estimate and final cost side by side |
| Using parts | Taken from the godown or bought; nobody records which | Each job-card part carries its source: stock (deducted automatically), purchased (raises an expense and a vendor bill), reused from another truck, or reconditioned |
| Knowing when a service is due | Diary, whiteboard, the driver’s memory | Schedule per truck and service type by km, days or both; a due board with overdue, due today and upcoming; next due recomputed when the service is recorded |
| Tracking a tyre | Tyre bills in a file; positions unknown | Tyre master with serial, brand, size and price; an axle layout per truck; 12 event types from purchase to scrap; tyre life and cost per km |
| Managing a battery | Replace when the truck will not start | Health, voltage and water level per inspection; warranty months; warranty claim events; weak-battery and warranty-ending alerts |
| Choosing a garage | Lowest estimate | Workshop comparison: jobs, total, average job, repeat-repair rate, estimate variance and a quality score per vendor |
| Approving driver expenses | WhatsApp photo, approved by habit | Driver submits from the phone; queue of pending approvals; only the owner role can approve; expense settles against the driver’s advance or goes to payables |
| Paying vendors | Ledger and reminders from the vendor | Every expense, job card, service and purchase creates a bill with a due date; payments from cash or bank; status paid, partial, unpaid, overdue; due and overdue notifications |
| Renewing documents | Discovered at a checkpost | 14 document types with expiry; notifications 30 days before a document and 45 days before a licence expires |
| Finding the costly truck | Gut feel | Maintenance cost per km per truck against the fleet average, a 0–100 health score with grade A to D, and an insight that names the truck with its evidence |
| Month-end review | The accountant’s totals | 25 reports and a monthly AI report as PDF: category breakdown, top-5 costly and most efficient trucks, driver comparison, tyre and battery analysis, pending payments, predicted upcoming expenses |
Want a fleet maintenance ERP built around your own trucks and garages?Send us a month of garage bills and your service diary for two trucks, and we’ll show you what one system would replace. Free workflow audit, reply the same day.
Chat on WhatsAppTrucks, the 12-tab truck profile and drivers
Everything in this ERP hangs off the truck. The truck master holds the registration, chassis and engine numbers, make and model, the axle configuration (6-wheel, 10-wheel, 12-wheel or 14-wheel), ownership as owned, financed or leased, the assigned driver, a monthly maintenance budget, the status (available, on trip, under maintenance, on contract) with its history, and the expiry dates for insurance, PUC, fitness, permit and road tax. The truck list shows each vehicle as a card with its status and the alerts that apply to it, and a Fleet Sheet view lays the whole fleet out as a spreadsheet with CSV and Excel export for anyone who still wants to work in Excel.
The 12-tab truck profile
Open any truck and you get twelve tabs on one page: Overview (status, driver, budget against spend, alerts, the next services due), Trips (every trip with km, load, customer and income), Expenses (trip and repair expenses by category), Tyres (the axle layout with the tyre at each position and its history), Batteries (fitted batteries with health, voltage and warranty), Maintenance (repair and maintenance expense lines), Job cards (open and closed repairs with stage and cost), Spare parts (parts issued to this truck), Documents (RC, insurance, PUC, fitness, permit, tax receipt, bills and photos with expiry), Payments (bills and payments that relate to the truck), Service (the service schedule and history for this truck) and AI (this truck’s health score, insights and predictions). This is the screen that answers “what does this truck cost me, and why?” without opening a second system.
A clickable 3D truck
The profile also carries a 3D truck scene, built with three.js, that maps the tyre positions and the main parts — cabin, engine, diesel tank, batteries, load body — so a tyre or a part can be reached by clicking the truck rather than reading a list. It is a navigation aid, not a sensor view: nothing on the 3D model comes from the vehicle itself.
Drivers
Each driver has a profile with licence number and expiry, assigned truck, experience, documents, assignment history and advances. The driver page shows the advances outstanding and the driver’s own AI insights, in particular his diesel efficiency compared with what the fleet achieves on the same routes and in the same truck class, so the comparison is fair. Licence expiry raises a notification 45 days ahead. For the driver’s own screens see the driver app below.
Job card software for truck repair: the 10-stage workflow
A job card is the record of one repair from complaint to payment. Searches for “job card software” mostly return systems for garages; this one is built for the fleet owner, the party who pays, so it is organised around approval, parts and the gap between the estimate and the bill. Every job card moves through ten stages, and the stage history records who moved it and when:
- Job card created — truck, complaint, priority, expected date, workshop or in-house.
- Vehicle inspection — the mechanic’s findings and the work items proposed, each with an estimate and labour.
- Work approved — the owner or maintenance manager approves the estimate; the change_status action is role-gated.
- Repair in progress — the truck is in the workshop; its status shows under maintenance on the dashboard.
- Parts required — a branch for jobs waiting on parts; each part line carries its source.
- Work completed — the final cost is entered against the estimate.
- Vehicle tested — a road test before the truck is released.
- Delivered — the truck is back with the driver; status returns to available.
- Bill received — the workshop bill is linked to the job card and becomes a vendor bill with a due date.
- Payment completed — the bill is paid from a cash or bank account; the job card closes.
Parts by source
Each part on a job card is tagged as from stock (deducted from spare parts inventory automatically), purchased (creates a repair expense and a vendor bill), reused from another truck, or reconditioned. That is how the inventory stays correct without a separate stock entry, and how a reconditioned alternator does not get charged as a new one.
Mechanics by category, estimate vs final
Mechanics are chosen by category — body, denter, electrician, welder, engine, tyre technician, general — so a job that needs a welder and an electrician records both. The estimate and the final cost sit on the same card, and the difference feeds the vendor comparison’s estimate variance. A job card that passes its expected date without being delivered raises an overdue notification, and repeat repairs of the same category on the same truck within 30 days become an AI insight. The board view shows every open job as a card in its stage column, which is the screenshot above.
The job-card pattern is one Yukti AI has built in several manufacturing systems too, such as the route-based production job cards in the corrugated box ERP; a repair job for a truck and a production job for a box share the same discipline of stages, parts and estimate against actual.
Service schedule software: 13 service types by km, by days or both
Preventive maintenance is cheap only if it happens. The schedule module holds, per truck and per service type, an interval by kilometres, by days, or by both with whichever comes first. The thirteen service types cover engine oil, gear oil, differential or crown oil, greasing, wheel greasing, air filter, oil filter, fuel filter, brake service, clutch service, coolant, battery service and general inspection, and the list is a setting, not code. When a service is recorded the system stores the km and date it was done, computes the next due km and date, posts the expense and, if an outside workshop did it, the vendor bill.
The service-due board splits every schedule into overdue, due today and upcoming, across the fleet or for one truck, and a calendar view shows the month. The odometer that drives the km side comes from trip start and end readings entered by the office or the driver, because there is no GPS in this version; when telematics is connected, the same schedule reads the live odometer instead. Overdue services feed the truck health score and raise notifications, and several overdue services across the fleet become a single “bulk overdue services” insight rather than twenty separate alerts.
The Momai build’s approach to preventive schedules, which proposes a job card before the service is overdue, is described in the Momai maintenance section; this build keeps schedule and job card as separate records so a service can be recorded in the yard in seconds without opening a ten-stage job.
Breakdowns, roadside repairs and downtime
A breakdown is the most expensive maintenance event a fleet has, because the truck stops earning while it is being fixed. The breakdown record captures the date and time, the location and highway, the odometer, the category of failure, the mechanic who attended, where the repair was done (roadside, nearest garage or towed to the yard), parts, labour and travel cost separately, the downtime hours, photos, and a status of reported, in repair or resolved. A driver can report a breakdown from the app, which creates the record with the truck already filled in.
The breakdowns screen totals downtime by truck and by month, groups failures by location and highway, and shows repeat patterns: the same truck, the same category, the same stretch of road. Each resolved breakdown feeds the breakdown report and the breakdown-history component of the truck health score, and open breakdowns appear in notifications until they are resolved. The AI layer’s breakdown-risk prediction reads this history together with overdue services and repeat repairs, which is how a records-based system can warn about a truck before it stops again.
Tyre management software for a truck fleet: axle positions, retreading, godown and scrap
Nobody on the search results for tyre management software shows a fleet owner an axle layout with a tyre at every position and that tyre’s life from purchase to scrap. This build does. Each truck’s axle configuration — 6-wheel, 10-wheel, 12-wheel or 14-wheel — draws a layout with named positions such as Front Left, Front Right, Rear 1 Left Outer, Rear 1 Left Inner, Rear 2 Right Outer and Spare, and every position shows the tyre fitted there.
The tyre master
A tyre is a record of its own: serial number, brand, size, radial or nylon, purchase price, vendor, warranty, tread depth at the last inspection and retread count. Its status is one of in use, in godown, at retreading, sold or scrapped, and it only changes through events.
12 event types
Purchased, installed, removed, transferred to another truck, sent for retreading, returned from retreading, puncture, rotated, stored in the godown, sold, scrapped and inspected. Each event carries the date, the truck and position where relevant, the odometer and the cost, so the system can total the kilometres a tyre has run across every truck and position it has been on, add its purchase, retread and puncture costs, subtract what the scrap dealer paid, and give a tyre cost per km. The analytics tab compares brands by life and punctures, lists tyres below the tread limit and shows the replacement trend.
Punctures from the driver’s phone
In the driver app the puncture tile shows the same axle layout; the driver taps the tyre that failed, enters the repair cost and attaches the bill. The puncture lands on the right tyre and the right truck as a trip expense in the “tyre puncture” category, pending approval. Four punctures on one tyre within 30 days become an insight, and the scrap sale of a tyre records the amount received against its original cost. A worked tyre cost per km is in the worked example below.
Truck battery tracking: health, voltage, water level and warranty claims
Batteries follow the same lifecycle discipline as tyres, with fewer positions and one extra idea: the warranty. Each battery has a serial, brand, capacity, price, vendor and warranty in months. Inspections record health as a percentage, voltage and water level. Ten event types cover purchase, installation, removal, transfer, inspection, repair, warranty claim, storage, sale and scrap. The system raises a weak-battery alert when health or voltage falls below the configured limit and a warranty-ending alert before the warranty months run out, so a failing battery is claimed rather than replaced. The battery lifecycle and battery replacement reports show cost, age and brand across the fleet, and battery condition is one of the seven components of the truck health score.
Spare parts stock software for a transport company, and scrap and disposal
A fleet’s godown is a small parts shop without a shopkeeper. The spare parts module gives it one. Parts are organised in thirteen categories (filters, belts, brake parts, clutch parts, electrical, lights and bulbs, bearings, suspension, engine parts, gearbox parts, body parts, lubricants and consumables, other) with a minimum quantity, a storage location, the usual vendor and a warranty where one applies. Seven stock transaction types move quantity: stock in from a purchase, stock out, issue to a job card, return from a job card, transfer, damaged and scrapped. The link to the job card is the important one: a part tagged “from stock” on a job card is deducted the moment the job card uses it, with no separate stock entry, so the shelf and the screen agree.
When a part reaches or falls below its minimum, the system raises a low-stock notification, and critical parts at or below minimum become an AI insight with the trucks that are likely to need them next. The spare parts consumption report shows which parts, which trucks and which months, and the inventory valuation report prices the godown at cost, which most fleets have never seen as a number.
Scrap and disposal
Old tyres, dead batteries, replaced parts and other items are sold to scrap dealers, and that money usually goes unrecorded. The scrap module records each sale with the item, the buyer, the amount received and the item’s original cost, which closes the loop on a tyre’s or battery’s lifetime cost and feeds the scrap sale report.
Vendor and mechanic comparison: repeat-repair rate, estimate variance and quality score
A transporter deals with more vendors than a shop does: diesel stations, tyre shops, battery dealers, parts suppliers, workshops, independent mechanics, electricians, denters, welders and scrap dealers. The vendor master uses those eleven types, and each vendor has contact details, a rating and a ledger of purchases, bills, payments and outstanding. The screen lists vendors with total purchases, outstanding amount and rating, which is the screenshot above.
Which garage is really cheaper?
The workshop comparison answers the question every owner argues about. For each workshop or mechanic it computes the number of jobs, the total spent, the average job value, the repeat-repair rate (how often a job on the same truck in the same category came back within the set window), the estimate variance (how far final bills land from estimates, on average) and a quality score that combines those with the rating. The cheapest garage by average job can be the most expensive garage by repeat rate, and the comparison shows both columns side by side, with the rows linked to the job cards behind them. The mechanic and workshop comparison and repair cost comparison reports put the same figures on paper for a vendor negotiation.
Which diesel pump is cheapest?
Diesel entries carry the pump, the litres and the rate, so the diesel rate comparison report ranks the pumps the fleet actually uses by average rate over the period, and the AI layer surfaces the cheapest regular pump as an insight with the rupee difference across the fleet’s monthly litres. It is a pump-rate comparison from the fleet’s own bills, not a fuel-card or sensor integration, which this version does not have.
Expenses, approvals, vendor bills, payments, driver advances and receivables
Every rupee a truck spends enters the system in one of two expense families. Trip expenses use ten categories: diesel (litres, rate and pump), toll, loading, unloading, tyre puncture, driver allowance (bhaththu), accident, fine and settlement, extra maintenance on trip, other (chanda, RTO, police) and electrical on trip. Repair and maintenance expenses use nineteen categories — tyre purchase, battery purchase, spare parts, labour, electrical, self-starter, dynamo and alternator, bulb and wiring, body work, denter, patti work, welding, painting, engine repair, gearbox repair, suspension, brake system, routine service and other — and each expense can have multiple lines, each carrying a parts amount and a labour amount separately. An expense is either paid immediately or goes to payables, can carry attachments, and has an approval status.
Approvals
Driver-submitted expenses land in an approvals queue. Only the owner role holds the approve action; the accountant can edit finance and the maintenance manager can edit maintenance, but neither can approve a driver’s diesel claim. An approved expense that was paid from the driver’s pocket settles against his advance; one that is unpaid becomes a vendor bill. Possible duplicate entries (same truck, same amount, same day) are flagged by the AI layer as an anomaly rather than silently blocked.
Bills, accounts and payments
A vendor bill is created from an expense, a job card, a service, a purchase, a tyre or battery purchase or a stock receipt, with a due date. Payments are recorded from a cash or bank account by cash, bank transfer, UPI or cheque through a record-payment modal, and a bill’s status moves between unpaid, partial, paid and overdue on its own. The pending bills, vendor payment and cash and bank reports show what is owed, to whom and from where, and overdue bills raise notifications and an insight. Freight receivables record what customers owe for completed trips, with an insight for receivables older than 30 days.
Driver advances
Advances given to a driver before a trip are recorded against him, and approved expenses settle against the advance; the balance stays on his profile until it is cleared. Pending advances across the fleet are an AI insight, and the driver-wise expense report shows what each driver spent by category.
This is an operational money ledger for the fleet, not a statutory book of accounts. The Tally question — push vouchers to Tally for the CA, or move the books in stages — is answered in our Tally integration guide and is scoped at go-live; there is no Tally connection in this version.
Trips, mining trip costing and contract rental
This is a maintenance-first system, but maintenance cost per km needs kilometres and a truck’s profit needs income, so trips are in it. A trip has one of four types — external freight, internal business for a group company, contract and own business — with from and to, a normalised route name used for fair comparison, start and end odometer readings, load in tons, customer, freight income and a status of planned, in progress or completed. Trip profitability is reported per trip and per truck: income less diesel, toll, driver allowance and other trip expenses, and cost per km.
Material trips with mining challans
A tipper fleet does not only carry; it often buys the load. Mining and aggregate trips carry material purchase lines: the material (river sand, aggregate, grit), quantity, rate, supplier, the mining challan number, the ISTP (inter-state transit pass) where it applies, and loading and unloading charges. The purchase becomes a cost of the trip, so the margin on a sand run is the sale less the material, the diesel and the loading, not the freight alone. This is the part of the build that fits the tipper and dumper fleets serving Gujarat’s quarries, crushers and river-sand leases.
Contracts and rental
Trucks placed with a client on monthly rental, per trip or per km have a contract record with the client, the rate, the deposit, the term and a status. The dashboard’s utilisation split shows how many trucks are on own-fleet work, external freight, internal business, contract rental, available or under maintenance, and idle trucks without a contract or trips become an insight with their missed contribution. Invoicing the contract client, e-invoice and e-way bill are not in this version; they are built in the Momai build’s billing module and can be added here.
Truck documents and expiry tracking: 14 document types
Every truck and driver document is a record with a type, a file and, where it applies, an expiry date. The fourteen types are RC, insurance, PUC, fitness certificate, permit, road tax receipt, repair bill, purchase bill, tyre invoice, battery invoice, job-card photo, before-repair photo, after-repair photo and other. Expiry dates for insurance, PUC, fitness, permit and road tax also sit on the truck master, and licence expiry on the driver, so the expiry tracker does not depend on somebody uploading the paper. Notifications fire 30 days before a document expires and 45 days before a driver’s licence does, and expiring documents are one of the seven prediction kinds in the AI layer, with the estimated renewal cost.
This is an expiry tracker, not a full document vault: there is no version history and no stepped reminder ladder with WhatsApp messages in this build. Both exist in the Momai build’s document vault (a 60/30/15/7/3/1-day ladder) and can be added to this system.
The owner dashboard, 25 fleet reports and notifications from live data
The dashboard opens on fleet KPIs: trucks by status, the utilisation split, month-to-date trip, maintenance and diesel spend, the expense category breakdown, open job cards, pending approvals, pending bills, upcoming services and the AI executive insights. Every number is read from the same analytics layer the reports use, so the dashboard and the reports never disagree. Twenty-five reports, each with a period filter and export to PDF, Excel or CSV, are grouped as follows:
| Group | Reports |
|---|---|
| Expenses | Truck-wise monthly expense; truck-wise yearly expense; trip expense |
| Maintenance | Maintenance expense; maintenance cost per km; breakdown report; scheduled service report; mechanic and workshop comparison; repair cost comparison; repeated parts failure |
| Fuel | Diesel consumption by truck and driver; diesel rate comparison by pump |
| Drivers | Driver-wise expense; driver fuel efficiency (route-adjusted) |
| Tyres and batteries | Tyre lifecycle; puncture frequency; battery lifecycle; battery replacement |
| Inventory | Spare parts consumption; scrap sale; inventory valuation |
| Finance | Pending bills; vendor payment; cash and bank |
| Profitability | Trip profitability by trip and by truck |
Notifications generated from the data, never stored
The notification centre is computed every time it opens, from the records as they are now, so it is never stale and never has to be “cleared” by a job that forgot to run. It covers document and licence expiry, services due and overdue, worn tyres, battery warranty and health, vendor bills due and overdue, pending approvals, overdue job cards, low stock, open breakdowns and unusually high expenses. Read and dismissed states are kept per user. A global search (Ctrl+K) finds trucks, drivers, job cards, vendors, tyres and parts from anywhere.
The driver app: diesel, expenses, punctures, breakdowns and checklists from the phone
Drivers are the source of most of a fleet’s raw records and the least likely users of a desktop screen, so the driver app is a phone-width set of large tiles for the driver’s own truck only. Add Diesel takes litres, rate and a bill photo. Trip Expense takes toll, loading, unloading and other costs with the category in Gujarati as well as English. Tyre Puncture shows the truck’s axle layout and the driver taps the tyre. Breakdown reports a stoppage with location, highway and photos. Checklist runs the pre-trip and post-trip inspection. My Trips lists his trips, and a history tile shows what he has submitted and whether it was approved.
Everything a driver submits is pending until the owner approves it, so the app cannot spend money; it records what was spent. The same design principle — an app for drivers who are not comfortable with technology, with the office watching every entry — runs through the Society Auto driver app and the Momai driver app. WhatsApp-based entry (a driver sends a diesel bill photo to a number and the system reads it) is not in this version; see what WhatsApp would add.
Four roles and six gated actions
The build defines four roles. Super Admin / Owner sees everything, approves expenses, uses the AI intelligence centre and manages settings. Admin / Accountant works in expenses, payments, bills, vendors, contracts, trips, reports and documents. Maintenance Manager works in job cards, services, breakdowns, tyres, batteries, spare parts, scrap, vendors and reports. Driver uses the mobile app for his own truck. Beyond module visibility, six actions are gated individually: approve, edit_finance, edit_maintenance, delete, change_status and manage_settings, and a roles and permissions matrix in settings shows who holds what.
One honest point: in this version roles are switched from the header and shape the screens in the application; there is no login. In production, authentication and authorisation must also be enforced on the server and the database, which is part of the go-live scope for any real deployment and is described under technology.
Not sure which of your trucks is the expensive one?Tell us how you track job cards, tyres and garage bills today and we’ll show you what one system would replace. Free workflow audit, reply the same day.
Get a free workflow auditA worked example: maintenance cost per km, tyre cost per km, the driver index and the truck health score
These four calculations are the ones a fleet owner asks about most, so here they are with the arithmetic shown. Every figure below is computed from the build’s rules on illustrative inputs; none is a client figure, a fleet average or a result. The weights and thresholds are the engine’s settings and can be changed per fleet.
1. Maintenance cost per km for one truck
The maintenance cost per km report takes everything spent on keeping a truck running in the period — repair and maintenance expense lines, job-card costs and service costs — and divides it by the kilometres the truck ran, taken from its trip start and end readings.
| Step | Working | Result |
|---|---|---|
| Kilometres in the month | Sum of (end km − start km) over the truck’s completed trips: 9,200 km | 9,200 km |
| Maintenance spend in the month | Job card (clutch) ₹24,800 + routine service ₹6,100 + two tyre punctures ₹900 + electrical ₹3,400 + brake parts from stock ₹6,200 | ₹41,400 |
| Maintenance cost per km | ₹41,400 ÷ 9,200 km | ₹4.50 per km |
| Fleet average (same period) | Fleet maintenance spend ÷ fleet km, say ₹6,20,000 ÷ 2,00,000 km | ₹3.10 per km |
| Insight rule | Truck above fleet average by more than the threshold (here 45% above) | “Maintenance cost per km above fleet average”, with the job cards as evidence |
2. Tyre cost per km across a tyre’s life
A tyre’s cost is not its purchase price. The tyre lifecycle report adds every rupee the tyre consumed and divides by every kilometre it ran, across all the trucks and positions it was fitted on.
| Step | Working | Result |
|---|---|---|
| Purchase | New radial tyre bought | ₹18,500 |
| First life | Installed at a rear position; removed at the tread limit after 46,000 km, with one puncture repair ₹350 | 46,000 km |
| Retread | Sent for retreading ₹4,200; returned; installed on another truck; retread count 1 | ₹4,200 |
| Second life | Removed after 22,000 km and stored in the godown; later scrapped | 22,000 km |
| Scrap sale | Sold to a scrap dealer | − ₹600 |
| Lifetime cost | 18,500 + 350 + 4,200 − 600 | ₹22,450 |
| Lifetime km | 46,000 + 22,000 | 68,000 km |
| Tyre cost per km | 22,450 ÷ 68,000 | ₹0.33 per km |
Multiply by the number of tyres on a 10-wheeler and you have the fleet’s tyre cost per km per truck, and the brand comparison runs the same arithmetic per brand. That is the number a tyre-shop negotiation should be about.
3. The route-adjusted driver efficiency index
A plain km per litre comparison punishes the driver on the hilly or heavily loaded route. The driver efficiency index compares each driver’s diesel use with what the fleet achieves on the same normalised routes in the same truck class, where 100 means par. If the fleet’s 10-wheelers average 3.9 km per litre on a given route and a driver’s trips on that route in a 10-wheeler average 3.54 km per litre, his index on that route is 3.54 ÷ 3.9 × 100 = 91; his overall index weights his routes by the kilometres driven on each. A driver below the fleet’s threshold becomes an insight with his trips as evidence; the best route-adjusted driver is named as the most efficient, and the UI says that the comparison is route-adjusted so it is not read as an accusation.
4. The truck health score
The score is 0 to 100 with a grade from A to D, built from seven weighted components, each scored from the truck’s own records and scaled to its weight:
| Component | Weight | What it reads | Illustrative score |
|---|---|---|---|
| Maintenance frequency | 20 | Repair events in the period against the fleet | 14 |
| Breakdown history | 20 | Breakdowns and downtime hours | 20 (none) |
| Upcoming and overdue service | 15 | Schedules overdue or due soon | 8 (one overdue) |
| Tyre condition | 15 | Tread below limit, punctures, tyres at retread | 12 |
| Battery condition | 10 | Health %, voltage, warranty status | 10 |
| Repair expense vs fleet per km | 10 | Maintenance cost per km against the fleet average | 6 (45% above) |
| Kilometre usage and age | 10 | Odometer and vehicle age | 7 |
| Total | 100 | 77 |
A total of 77 for this illustrative truck places it in the middle grades: no breakdowns and a healthy battery, pulled down by one overdue service, a clutch job and a cost per km well above the fleet. The score’s value is the breakdown, not the number: the AI tab on the truck lists each component with the records behind it and the action that would lift it (record the overdue service, review the repeat clutch work with the workshop). Grade bands are an engine setting. See what a truck health score is for how it is meant to be read.
AI for a truck fleet: rule-based intelligence that shows its evidence
Say it plainly first: the AI in this build is rule-based intelligence on the fleet’s own records. There is no language model in it and no external AI service is connected. A deterministic engine of roughly 940 lines reads the same records every screen reads and returns findings, each with a plain finding, evidence lines, why it matters, an action, an estimated monthly rupee impact and a severity. It never invents a number, because every number it shows was computed from a record it links to. A language model can be added behind the same provider interface later, for natural-language questions and report writing, and the engine’s facts would be its only source. What that honest kind of AI looks like in general is described on our ERP-connected AI agents page.
The AI intelligence centre: nine tabs
- Business overview — an executive summary, what needs attention, where the money goes, costly versus efficient trucks, for 30, 90 or 180 days or this financial year.
- All insights — every finding from the 20 rule families, filterable by severity and truck.
- Anomalies — trip diesel against the route benchmark, job-card final bill against estimate, and possible duplicate entries.
- Predictive maintenance — seven prediction kinds: oil change, service, tyre replacement, battery, part failure, breakdown risk and document expiry, each with reason, history, due date or km, estimated cost and a confidence value, labelled as an estimate.
- Cost optimisation — the actions with the largest estimated monthly impact, with the arithmetic.
- Driver analysis — the route-adjusted fair comparison described above.
- Fleet health — the 0–100 score and A–D grade per truck with the seven components.
- Ask AI — questions in English or transliterated Gujarati, answered with figures, tables, charts and links.
- Monthly report — a management PDF.
The 20 insight families
Maintenance cost per km above the fleet average; diesel km per litre below the benchmark for the same axle class; a driver’s trip expenses above the average driver; punctures in 30 days; oil service due or overdue; bulk overdue services; tyre replacement trend; cheapest regular diesel pump; workshop repeat-repair rate; overdue vendor bills; document expiry; possible duplicate expense entry; monthly budget overrun per truck; idle trucks and their missed contribution; freight receivables older than 30 days; unusually high miscellaneous driver expenses; the best fuel-efficiency driver, route-adjusted; weak batteries; critical spare parts at or below minimum; repeat repairs of the same category within 30 days. Each insight links to the truck tab, job card, tyre, vendor or report it came from.
Ask AI in Gujlish
The chat maps questions to intent handlers over the computed data, so an owner can type “kaya truck no kharch sauthi vadhu?” (which truck costs the most?) or “service due aavta 15 divas ma” (services due in the next 15 days) as easily as the English equivalents, and get a table, a chart and links with follow-up suggestions. It answers only from computed data; a question it has no handler for gets a polite “I can’t answer that yet”, not a guess. This is the part a language model would extend most, by understanding free-form questions, while the engine keeps supplying the facts.
The monthly report
One click produces a PDF with an executive summary, the category breakdown, the top-5 costly and most efficient trucks, the driver comparison, tyre and battery analysis, the maintenance trend, pending payments, predicted upcoming expenses and cost-saving recommendations. It is the document an owner can take to the accountant, the workshop or the bank. The AI narration pattern in the Momai build and the voice and WhatsApp agents on our AI automation agency page follow the same rule: AI explains computed facts and never originates a figure.
It is not a digital twin, not sensor-based predictive maintenance, not machine learning trained on the fleet, and not a chatbot that writes prose. The “savings opportunity” figures it shows are estimates computed from the records in front of it, and the figures visible in the screenshots come from the sample dataset, not from the client’s fleet.
What is a truck health score, and how should a fleet owner read it?
A truck health score is a single number, here from 0 to 100 with a grade from A to D, that summarises how much attention a truck needs, computed from its maintenance frequency, breakdown history, overdue and upcoming services, tyre and battery condition, repair spend per km against the fleet and its kilometres and age. It is a ranking tool, not a diagnosis: it tells the owner which truck to look at first and shows the seven components so the reason is visible. In the Rajeev Roadlines build the score is recomputed from the records whenever a screen opens, lives on the truck’s AI tab and on the Fleet Health tab of the AI centre, and every component links to the job cards, services, tyres and batteries behind it. A score built from sensor data, as hardware telematics vendors offer, measures the engine; a score built from records, as this one does, measures how the truck is being managed. The two are complementary, and the records-based score is the one a fleet can have without fitting any device.
AI and automation for a truck fleet: built here, built in other Yukti systems, buildable, and not built
Fleet software marketing blurs four very different things: what runs in the system you are reading about, what the same company has built elsewhere, what could be built, and what needs hardware or a third-party provider. This table keeps them apart. “Built here” means it exists in the Rajeev Roadlines system. “Built in another Yukti system” links to the case study where it runs. “Can be built” is scoped with the client. “Needs a provider or sensors” cannot be done by software alone.
| Capability | Status | Notes |
|---|---|---|
| Notifications from live data (expiry, service due, worn tyres, battery, bills, approvals, overdue job cards, low stock, breakdowns, high expenses) | Built here | Computed on open from the records; read and dismissed state per user |
| Job-card parts deducting from stock; purchased parts raising an expense and a vendor bill | Built here | By part source on the job card |
| Service next-due recomputed on record; due board | Built here | By km, days or both |
| Bills created from expenses, job cards, services and purchases; bill status from payments | Built here | Cash, bank, UPI, cheque |
| Tyre and battery bookkeeping on install, transfer, retread, sale and scrap | Built here | 12 tyre and 10 battery event types |
| Driver expense approval queue with owner-only approve | Built here | Settles against advances or goes to payables |
| Rule-based insights, anomalies, predictions, health score, driver index, Gujlish chat, monthly PDF | Built here | No language model; LLM can be added behind the same interface |
| Automation rule engine with trigger, condition, action and an activity log | Built in another Yukti system | Momai fleet ERP: 16 rules, autonomous or propose-only; can be added here |
| Diesel bill OCR with a “review required” queue | Built in another Yukti system | Momai fuel module (adapter awaiting its provider) |
| Document vault with versions and a 60/30/15/7/3/1-day reminder ladder | Built in another Yukti system | Momai document vault |
| EMI and bank statement mapping per truck | Built in another Yukti system | Momai money module |
| Invoices, e-invoice and e-way bill screens for transport contracts | Built in another Yukti system | Momai billing; providers connect at go-live |
| Voice AI payment reminder calls to customers who owe freight | Built in another Yukti system | Sureja voice AI reminders; see voice AI payment reminder calls |
| WhatsApp bots and template messages from ERP events | Built in other Yukti systems | WhatsApp automation with the WhatsApp Business API; see API pricing |
| WhatsApp alerts for service due, document expiry, bill due and job-card status | Can be built | From this system’s notifications, through approved templates |
| Driver submits diesel, puncture or breakdown by WhatsApp instead of the app | Can be built | Photo and text into the same approval queue |
| Tally sync of vendor bills, payments and expenses | Can be built | See Tally integration with a custom ERP |
| GST invoicing, e-invoice, e-way bill, LR / bilty | Can be built | Pattern exists in the Momai build; GST providers connect at go-live |
| Language model for free-form questions and report prose | Can be built | Behind the existing provider seam; engine facts as the only source |
| Server database, login and server-side permissions | Can be built | Entities are typed one-to-one with future tables; a repository interface is in place |
| Live GPS / telematics, AIS-140 trackers, live odometer | Needs a provider | Listed as planned in settings; odometer entered manually today |
| FASTag toll statement import | Needs a provider | Planned; toll entered manually today |
| Fuel card or pump statement import | Needs a provider | Planned; diesel entered per fill today |
| Fuel-level sensors, tyre-pressure monitoring (TPMS), OBD / CAN engine data | Needs sensors | Would improve the predictions; not required for them |
| Dashcams, driver-behaviour video, ADAS | Needs hardware | Not in scope for a maintenance ERP |
| VAHAN and SARATHI lookups, load marketplace | Not built | Third-party services; can be scoped |
The cost of the automation and AI additions follows the same rule as the ERP: scoped after the audit, with the published starting points in our AI automation cost guide.
What WhatsApp, GPS, FASTag, fuel cards and Tally would add to this fleet ERP
None of these is connected in the version described on this page. Each is a bounded addition that fits a seam already in the build, and this is what each would change for the fleet.
WhatsApp alerts and driver entry
The notification centre already knows when a service is due, a permit is expiring, a vendor bill is overdue or a job card is late. Sending those as WhatsApp messages to the owner, the maintenance manager or the driver needs a WhatsApp Business API number and approved message templates, and then the same alerts arrive on the phone instead of waiting for someone to open the system. In the other direction, a driver can send a diesel bill photo, a puncture note or a breakdown message to the business number and have it land in the approval queue. Yukti AI builds this on the WhatsApp Business API; see WhatsApp automation for what is involved and WhatsApp API pricing in India for what Meta charges per message.
GPS and telematics
With GPS the odometer that drives service-due-by-km becomes live, trip start and end km are captured without typing, breakdown locations are exact and idle time is visible. Settings in the build list GPS as planned; connecting it means one adapter for the chosen provider’s API. GPS does not replace the maintenance records — a tracker does not know a clutch was changed — which is why the comparison below treats GPS tracking apps and a maintenance ERP as different tools.
FASTag and fuel cards
A FASTag statement import turns toll from a driver claim into a bank-verified line per trip, and a fuel card ledger does the same for diesel at participating pumps. Both are planned in settings and both are statement imports from the issuing bank or oil company. Until then, toll and diesel are entered per trip and checked by the approval queue and the AI anomaly rules.
Tally
The fleet’s accountant and CA usually keep the statutory books in Tally. The ERP’s vendor bills, payments, expenses and receivables can be pushed as vouchers so the books stay in Tally while the operational truth stays in the ERP, or the fleet can move its books in stages. Which is right depends on the CA and on GST filing habits, and it is scoped at go-live; the trade-offs are on our Tally integration page.
Voice AI for collections
Freight receivables older than 30 days are already an insight. A voice AI agent that calls the customer’s accounts desk with the invoice number and amount, logs the promise to pay and escalates to a person is something Yukti AI has built for a Surat business; read the Sureja payment reminder voice AI case study.
Custom fleet maintenance ERP vs GPS tracking apps vs ready-made fleet software vs Excel and diaries
A transporter searching for fleet software in India meets four kinds of answer. This is a general comparison of the categories, not a review of any product, and the competitor features it refers to are the vendors’ own claims.
| Question | Excel + diaries | GPS tracking app | Ready-made fleet software | Custom maintenance ERP (as built here) |
|---|---|---|---|---|
| Where is the truck? | Phone the driver | Live, with history and geofences | Usually via a tracker | Not in this version; GPS connects as an addition |
| What did this truck cost to maintain per km? | Not known | Not known; service reminders at best | Sometimes, if every bill is entered | Yes, per truck and against the fleet, from job cards, services, tyres, batteries and parts |
| Job cards with stages and parts | No | No | Varies; often a workshop module | 10 stages, parts by source, mechanics by category, estimate vs final |
| Tyre and battery lifecycle | Bills in a file | TPMS alerts with sensors | Tyre stock in some products | Axle positions, 12 tyre and 10 battery events, retread, godown, scrap, cost per km |
| Which garage is cheaper? | Opinion | No | Rarely | Repeat-repair rate, estimate variance, quality score |
| Predictive maintenance | No | With OBD / sensors, from engine data | Reminders | From records: due km and days, wear, repeat repairs, breakdown risk, labelled estimates; sensors improve it |
| Driver expense control | WhatsApp photos | Fuel sensors in some | Expense entry | Driver app, approval queue, route-adjusted comparison, anomaly rules |
| GST billing, LR, e-way bill | Tally | No | Often included in transport ERPs | Not in this version; built in the Momai system and can be added |
| Fits your fleet’s categories, roles and approvals | Yes, by hand | No | Partly | Yes: your categories, intervals, garages, roles and approval rules |
| AI | No | Driver scoring, alerts | Rarely, and usually sensor-led | Rule-based insights with evidence, health score, Gujlish chat; LLM when you want it |
| Cost model | Cheap, costly in mistakes | Per vehicle per month plus device | Licence or per-vehicle subscription | One-time build with support; from ₹30,000, quoted after a free audit |
| Time to start | Immediate | Device fitting | Implementation weeks | First working version in about 7–10 days; full build a few weeks |
The short version: a GPS app and a maintenance ERP answer different questions, and most fleets of 20 trucks and more eventually want both. For the general argument see custom ERP vs off-the-shelf ERP.
Features of truck maintenance ERP software, and which ones this build covers
Fleet and maintenance products describe themselves in the same groups: vehicles and documents, maintenance and workshop, tyres and parts, fuel and drivers, finance, reports and AI, and integrations. This checklist uses those groups so you can compare any product, including a custom build, feature by feature. “Covered” means it exists in the Rajeev Roadlines system; “Can be added” means it is not built and would be scoped with the client; “Needs a provider or sensors” means software alone cannot do it.
Vehicles, drivers and documents
| Feature | What it does | In this build |
|---|---|---|
| Truck master with 12-tab profile | RC, chassis, engine, ownership, axle configuration, budget, status history, every record of the truck on one page | Covered |
| Fleet sheet with CSV / Excel export | Spreadsheet view of the fleet | Covered |
| Clickable 3D truck | Navigation by tyre position and part | Covered |
| Driver profiles with licence expiry, advances and assignment history | Driver master and ledger | Covered |
| 14 document types with expiry notifications | 30 days for documents, 45 for licences | Covered |
| Document versions and reminder ladder | Stepped reminders with history | Can be added (built in the Momai system) |
| VAHAN / SARATHI lookups | RC, challan, PUC, fitness, permit checks | Needs a provider |
Maintenance and workshop
| Feature | What it does | In this build |
|---|---|---|
| 10-stage job cards | Complaint to payment, with stage history | Covered |
| Parts by source on the job card | Stock, purchased, reused, reconditioned | Covered |
| Mechanics by category | Body, denter, electrician, welder, engine, tyre, general | Covered |
| Estimate vs final cost | On the card and in the vendor comparison | Covered |
| 13 service types by km, days or both | Next due computed on record; due board; calendar | Covered |
| Breakdowns with highway, mechanic, costs and downtime hours | Reported from the driver app; repeat patterns | Covered |
| Pre and post-trip checklist | In the driver app | Covered |
| Workshop portal for outside garages | Garage updates the job card itself | Can be added |
Tyres, batteries and parts
| Feature | What it does | In this build |
|---|---|---|
| Axle layouts for 6W, 10W, 12W and 14W | Named positions with the tyre fitted | Covered |
| 12 tyre event types incl. retread, godown, sale, scrap | Full lifecycle; tyre life and cost per km; brand comparison | Covered |
| Puncture from the driver’s phone by tapping the tyre | Lands on the right tyre, pending approval | Covered |
| Battery health, voltage, water level, warranty claims | 10 event types; weak and warranty-ending alerts | Covered |
| Spare parts in 13 categories with minimum stock | 7 transaction types; job-card deduction; valuation | Covered |
| Scrap and disposal sales | Buyer, amount received, original cost | Covered |
| Tyre pressure monitoring | Live pressure and temperature | Needs sensors |
| Purchase orders to parts suppliers | PO, receipt, bill matching | Can be added |
Fuel, drivers and trips
| Feature | What it does | In this build |
|---|---|---|
| Diesel per fill with litres, rate and pump | Km per litre by truck and driver; pump rate comparison | Covered |
| Route-adjusted driver efficiency index | Fair comparison on the same routes and truck class | Covered |
| Driver app for diesel, expenses, punctures, breakdowns, checklists | Own truck only; everything pending approval | Covered |
| Trips in 4 types with material purchase lines | Mining challan, ISTP, loading and unloading; trip profitability | Covered |
| Contracts: monthly, per trip, per km | Client, rate, deposit, term | Covered |
| Diesel bill OCR with review queue | Reads the bill photo | Can be added (built in the Momai system) |
| Fuel-level sensors, fuel cards, FASTag import | Verified diesel and toll | Needs a provider or sensors |
| Live GPS tracking | Location, route, live odometer | Needs a provider |
Finance, reports and AI
| Feature | What it does | In this build |
|---|---|---|
| 10 trip and 19 repair expense categories, multi-line | Parts and labour separately | Covered |
| Approvals queue, owner-only approve | Driver claims controlled | Covered |
| Vendor bills, cash and bank accounts, payments, advances, receivables | Operational ledger | Covered |
| 11 vendor types with workshop comparison | Repeat-repair rate, estimate variance, quality score | Covered |
| 25 reports with PDF / Excel / CSV | Expenses, maintenance, fuel, drivers, tyres and batteries, inventory, finance, profitability | Covered |
| Rule-based AI: insights, anomalies, predictions, health score, chat, monthly PDF | Evidence-backed; no language model | Covered |
| Language model on top | Free-form questions and report prose | Can be added |
| GST invoicing, e-invoice, e-way bill, LR | Freight billing | Can be added (built in the Momai system) |
| Tally sync | Vouchers to the CA’s books | Can be added |
| WhatsApp alerts and driver entry | Service, expiry, bill and job-card messages | Can be added |
| Server database, login, server-side roles | Multi-device, multi-user production use | Can be added (go-live scope) |
Benefits of fleet maintenance software for a transporter, feature by feature
Benefits here are the kind of result each feature is designed to make possible, not measured outcomes from this client. No percentage, rupee saving or downtime reduction is claimed on this page.
| Feature | What it makes possible |
|---|---|
| Every rupee attached to a truck and a kilometre | Maintenance cost per km per truck, so the expensive truck is a fact, and the decision to repair, re-deploy or sell it has a number behind it |
| Service schedule by km and days | Services happen when due rather than when remembered; the due board replaces the diary and the whiteboard |
| 10-stage job cards | Nobody pays a garage bill for work that was not approved, and the estimate-to-bill gap is visible per workshop |
| Tyre lifecycle with retreading and scrap | Tyre cost per km by brand, punctures per tyre and a case for or against retreading from the fleet’s own numbers |
| Battery warranty tracking | Claims made before the warranty ends instead of batteries bought twice |
| Parts deducted by job cards | Godown stock that matches the shelf; fewer duplicate purchases; a valued inventory |
| Workshop comparison | Garage choice by repeat-repair rate and estimate variance, not by the lowest quote |
| Approval queue and route-adjusted driver index | Driver claims checked against the fleet on the same route; a fair comparison that drivers can see the logic of |
| Notifications from live data | Expiries, bills, services and low stock surface before they cost a fine, a stoppage or a penalty |
| Rule-based AI with evidence | An owner reads findings and actions instead of 25 truck ledgers; every finding can be checked in one click |
| Gujlish chat | The owner asks in his own words and gets a table, not a training course |
| Monthly PDF | One document for the accountant, the workshop negotiation and the bank |
| Driver app | Records entered at the point of spending by the person who spent, with the office in control of approval |
Buying checklist: 12 questions to ask before you choose truck maintenance software
- Does it attach every expense to a truck and a kilometre? If maintenance cost per km per truck is not a standard report, the system is a diary with a login.
- Does the job card have stages, parts by source and estimate vs final? A job card that is a single form cannot tell you which garage overruns its estimates.
- Can service intervals be by km, by days or whichever comes first, per truck and per service type? Trucks on short city runs and long highway runs need different rules.
- Does it show tyres by axle position and follow a tyre through retreading, the godown and scrap? Tyre stock alone does not give tyre cost per km.
- Are batteries tracked with warranty months and health readings?
- Do parts used on a job card come off the stock automatically? If not, stock will be wrong within a month.
- Can it compare workshops by repeat-repair rate, not only by spend?
- Is the driver comparison route-adjusted? A plain km-per-litre league table is unfair and drivers know it.
- Where does the AI get its numbers? Ask for the evidence behind one insight. If the vendor cannot show it, it is a slogan.
- Which integrations are live today, and which are “planned”? GPS, FASTag, fuel cards, WhatsApp and Tally should each be answered with a yes, a no or a date.
- Can your accountant keep Tally? Decide the books question before the build, not after.
- Who owns the data and the code, and what does support cost after go-live? A custom build should answer: you, and a fixed support arrangement.
The same questions, asked about this build, are answered on this page: see the feature checklist and what this page does not claim.
Truck maintenance software for Surat and Gujarat fleets: who this is for, and how we work
Surat’s trucking runs on a few corridors: NH-48 north towards Bharuch, Vadodara and Ahmedabad and south towards Vapi, Mumbai and Bhiwandi; NH-53 east towards Bardoli, Songadh and the Maharashtra border; the port and industrial traffic around Hazira; the textile and diamond freight out of Sachin GIDC, Pandesara, Udhna and Kadodara; and the tipper trade between the river-sand and aggregate leases and the construction sites, with yards strung along Kamrej, Kadodara, Palsana and Kosamba. Those are general facts about Surat’s geography, not statements about the client. A fleet on those roads has the same maintenance problem whether it carries fabric, chemicals, containers or sand: the trucks are serviced at yards and roadside garages between runs, tyres are retreaded at local retreaders, and the expenses come home as bills, photos and cash.
Yukti AI builds for transporters in Surat, across Gujarat (Ahmedabad, Vadodara, Rajkot, Bharuch, Ankleshwar, Vapi, Navsari, Bhavnagar, Morbi) and across India. The audit is in Hindi, Gujarati or English, on WhatsApp, on a call or at our office at 220, Leonard Square, Yogichowk, Varachha, Surat – 395006. We work from your real categories: your garages, your service intervals, your axle configurations, your expense heads with their Gujarati names, and your approval habits. GJ-series registration, Gujarat RTO permits, state and national permits, PUC and fitness cycles are the document set the expiry tracker is built around.
For the broader picture of what we do for transporters, see fleet management software and logistics and transport; for the ERP practice in general, custom ERP software and custom ERP software in Surat; and for the company, about Yukti AI.
One dataset, one analytics layer, one AI engine
Every screen, report, notification and AI insight reads the same records through the same analytics layer, so the dashboard, the truck profile, the reports and the health score cannot disagree. Integrations sit behind seams that are already in the code.
The 22 modules
Twenty-two modules in the sidebar on one shared set of records — the scope of a full AI-powered business management system, shaped around a truck fleet’s maintenance — plus the driver app.
Overview
Dashboard; AI Intelligence Center; Fleet Sheet.
Fleet operations
Trucks with the 12-tab profile and 3D scene; Drivers; Trips with trip costing and material purchase lines; Contracts & Rental.
Expenses
Trip Expenses (10 categories); Repair & Maintenance (19 categories, multi-line); Approvals.
Maintenance
Job Cards (10 stages); Service Schedule (13 types); Breakdowns.
Assets and inventory
Tyres (axle layouts, 12 events); Batteries (10 events); Spare Parts (13 categories, 7 transaction types); Scrap & Disposal.
Finance
Payments & Bills (bills, cash and bank accounts, payments, driver advances, receivables); Vendors & Mechanics (11 types, workshop comparison).
Insights and administration
Reports (25); Documents (14 types, expiry tracker); Notifications; Settings (company profile, truck budgets, data, roles and permissions matrix, integrations status); global search.
Driver app
Add Diesel, Trip Expense, Tyre Puncture, Breakdown, Checklist, My Trips, history.
How it was built
Start from the truck
We modelled the truck as the asset every record hangs off, with its tabs defined before any list screen, because “what does this truck cost me” is the question the system exists to answer.
One analytics layer
Cost per km, service due, tyre life, workshop comparison and trip profit were written once as selectors, so the dashboard, the reports and the AI read the same arithmetic.
Events, not edits
Tyres, batteries and stock change state only through typed events (installed, retreaded, scrapped, issued), so a tyre’s history is a ledger rather than a field somebody overwrote.
Rules with evidence
Each AI rule was written to return its evidence lines, its rupee estimate and a link to the record, and to say nothing when the data does not support a finding.
Realistic records
The records the system ships with were generated from a fixed seed over about eighteen months of history, so every total on every screen and report agrees and the owner can judge a working system.
Settings, not code
Service types and intervals, expense categories, part categories, vendor types, truck budgets, roles and the integrations list are settings the fleet can change itself.
See your own trucks, garages and tyres in a working system firstWe build on your real service intervals, expense heads and workshops, so you judge something working, not a slide deck. Free workflow audit, reply the same day.
Chat on WhatsAppWhat was delivered
What the system does — no invented numbers.
What this page does not claim
- No claim that Rajeev Roadlines uses the system in daily operation, how many people use it or since when. There is no client quote and no outcome figure. The fleet size of 25 trucks is the figure the system was built for, per the brief, and is not independently verified here.
- No claim of money saved, breakdowns reduced, downtime cut, tyre life extended or services done on time. The rupee estimates and “savings opportunity” figures the AI centre shows in the screenshots are computed from the sample dataset the system ships with and are not the company’s business numbers.
- Truck registration numbers, driver, garage, vendor, pump and customer names visible in the screenshots are from the sample dataset, not the client’s records.
- No live connections: GPS or telematics, FASTag, fuel cards, WhatsApp, Tally, e-invoice, e-way bill, bank feeds, sensors and a server database are not connected in this version, and there is no language model or machine-learning feature. Settings list GPS, FASTag and fuel cards as planned.
- Records are kept in the browser on the device that uses the system and survive reloads; there is no login, and roles are switched from the header. Multi-user production use needs the server database and authentication described under technology.
- The money module is an operational ledger, not a statutory book of accounts or a GST return, and there is no freight invoicing, LR or e-way bill in this build.
- The worked example uses illustrative inputs to show the build’s arithmetic; the figures are not fleet averages or client results, and the engine’s weights and thresholds are settings.
- No document version history or reminder ladder, no automation rule engine with a scheduler, no load marketplace and no dashcam or sensor integration in this build; each is listed above as an addition.
The working version built by Yukti AI around the fleet’s workflow, with records kept in the browser. Anything not listed as built above is described as buildable, not as delivered.
Technology
The build uses React 18 with TypeScript 5 and Vite 5, Tailwind CSS v4, React Router 6, Zustand with Immer for the store, idb-keyval for IndexedDB persistence in the browser, Recharts for charts, lucide-react for icons, three.js with react-three-fiber for the 3D truck scene, jsPDF with autotable for PDF reports and SheetJS for Excel and CSV export. It is configured for static hosting with a deploy workflow. Every entity is typed one-to-one with a future database table and data access goes through a repository interface with a local implementation, so adding a server database means implementing that interface, not rewriting screens. The AI sits behind a provider interface whose active provider is the local rules engine; a language-model provider can be registered behind the same interface. GPS, FASTag and fuel cards appear in the settings integration list as planned connections.
Going live for daily, multi-user operation means adding the server database with authentication and server-side enforcement of the four roles and six actions, migrating the fleet’s trucks, tyres, batteries, parts, vendors and open bills, connecting the chosen GPS, FASTag, fuel-card, WhatsApp and Tally providers, and training the office and drivers. None of that is in the version this page describes.
How much does a truck maintenance ERP cost in Surat, and how long does it take?
A custom ERP from Yukti AI starts from ₹30,000. That starting point covers a focused system — one clear job done well, such as job cards with a service schedule and tyre tracking for a small fleet. A multi-module fleet ERP of the breadth described here, with trips, expenses, approvals, tyres, batteries, parts, vendors, bills, documents, 25 reports, a driver app and the AI layer, costs more, and the final figure is a fixed quote agreed after a free 30-minute workflow audit. There is no per-user or per-vehicle licence. Adding GPS, FASTag, fuel-card, WhatsApp or Tally connections, a server database with login, or a language model on top of the rules engine each adds scope and is priced in the same quote.
On timing: a first working version is typically ready in about 7–10 days, built on your real service intervals, categories and garages so you can judge something that runs. A multi-module ERP usually takes a few weeks depending on scope, and a full rollout with data migration, server-side roles, integrations and training follows from there. For what drives the price in general see custom ERP software cost in India, and for the automation and AI side AI automation cost in India. Public price anchors for ready-made transport ERPs and per-vehicle fleet subscriptions are listed on third-party marketplaces and vendor sites; they are their claims, and a custom build is priced on scope rather than per vehicle.
The free audit is a 30-minute conversation about how job cards, services, tyres, garage bills and driver expenses are handled today and where the leaks are. Clear scope, no obligation. Start on the contact page.
Which Yukti AI fleet build fits your transport business?
Yukti AI has built two transport systems for Surat fleets, with different centres. One fleet can have both in one system; most start with the side where the money leaks most.
| Shree Momai fleet management ERP | Rajeev Roadlines truck maintenance ERP (this page) | |
|---|---|---|
| Centre | The trip and the money: trips with photo evidence, diesel OCR review, document vault with reminder ladder, EMI and bank mapping, contracts, invoices with e-invoice and e-way bill screens, 16 automation rules, 7 roles, 39 reports | The truck as an asset: 10-stage job cards, service schedule, breakdowns, tyre and battery lifecycle, spare parts stock, scrap, vendor and mechanic comparison, truck health score, predictive maintenance from records, 4 roles, 25 reports |
| Fleet in the build | 12 trucks, 15 drivers, 120 days | 25 trucks, 22 drivers, about 18 months of history |
| AI | Deterministic narration of computed facts; language model can be swapped in | Deterministic rules and statistics; health score; Gujlish chat; monthly PDF; language model can be swapped in |
| Integrations | 10 adapters awaiting providers (GPS, WhatsApp, accounting, e-invoice, e-way bill, bank, OCR, FASTag, notifications, AI) | None connected; GPS, FASTag and fuel cards planned in settings |
| Persistence | Session only; reload resets | IndexedDB in the browser; survives reload on the device |
| Unique to it | Trip evidence, fuel anomaly OCR, document versions and ladder, EMI mapping, invoicing, automation engine | 3D truck with clickable tyres, retreading and godown flow, battery health and warranty, parts-to-job-card stock, scrap sales, driver advances, mining and material trip costing, route-adjusted driver index, Gujarati labels |
| Read it | Momai fleet ERP case study | This page; the service overview is on fleet management software |
Truck maintenance ERP and fleet software: questions transporters ask
Explore more
Fleet Management Software
Transport ERP for truck owners in Surat, Gujarat and India: trips, fuel, maintenance, documents, billing.
Case study: Shree Momai fleet ERP
Trips with evidence, diesel OCR, document ladder, EMIs, invoicing and 16 automation rules.
AI for Logistics & Transport
Where AI and automation help Indian transporters today.
Custom ERP Software in Surat
ERP built around how your Gujarat business actually runs.
Custom ERP Software
One system for operations, stock, money and reports, built on your workflow.
WhatsApp Automation
Service-due, expiry and bill alerts and driver entry on the WhatsApp Business API.
Voice AI Payment Reminders
AI calls that chase freight receivables and log the promise to pay.
Case study: Sureja voice AI
Payment reminder voice AI built for a Surat business.
ERP-Connected AI Agents
AI that reads your data, shows its evidence and never invents a number.
AI Automation Agency in Surat
WhatsApp, voice, n8n and AI automation for Surat businesses.
Tally Integration
Connect a custom ERP to Tally, or move off it in stages.
Custom ERP Cost in India
What drives the price of a custom ERP and how to budget.
All case studies
Every Yukti AI build across transport, manufacturing, diamond, jewellery and trading.
Summary
Yukti AI, a software company in Surat, built a truck maintenance ERP for Rajeev Roadlines, a Surat transporter with a fleet of 25 trucks (per the brief) across 6-, 10- and 12-wheel configurations running external freight, group-company haulage, contract rental and material trips. The system treats the truck as an asset: a 12-tab truck profile with a clickable 3D truck; drivers with licences and advances; trips in four types with material purchase lines for mining and aggregate runs; trip expenses in 10 categories and repair expenses in 19 with parts and labour separated; an approvals queue where only the owner approves; job cards through 10 stages with parts by source (stock, purchased, reused, reconditioned) that deduct from inventory; a service schedule of 13 types by km, days or both with a due board; breakdowns with highway, mechanic, costs and downtime; tyres by axle position with 12 lifecycle events through retreading, godown, sale and scrap and a tyre cost per km; batteries with health, voltage, water level and warranty claims; spare parts in 13 categories with 7 transaction types and valuation; scrap sales; vendor bills, cash and bank accounts, payments, driver advances and receivables; a vendor comparison by repeat-repair rate, estimate variance and quality score; contracts; 14 document types with expiry notifications; 25 reports with PDF, Excel and CSV; notifications computed from live data; a driver app; and four roles with six gated actions. The AI layer is rule-based intelligence on the fleet’s own records: 20 insight families, anomalies, 7 prediction kinds, a 7-component truck health score from 0 to 100 with grades A to D, a route-adjusted driver efficiency index, chat in English and transliterated Gujarati and a monthly PDF; no language model is connected and one can be added behind the same interface. It is the working version built around the fleet’s workflow with records kept in the browser; GPS, FASTag, fuel cards, WhatsApp, Tally, GST billing, sensors and a server database are not connected and are listed as additions; usage and results are not claimed. A custom ERP from Yukti AI starts from ₹30,000 with a first working version in about 7–10 days and a quote after a free workflow audit.