Job Shop Work Order Tracker for Small Manufacturers

The problem small manufacturers have with tracking jobs

Small manufacturers such as steel fabricators, furniture workshops, print shops and garment stitching units work job by job. Each order is a bit different. The owner quotes a price from experience, writes a job card, and then chases the floor all day to find out where things are. Nobody knows if a job made money until the month ends, and often not even then. When a big customer asks "when will my order be ready?", the answer is a guess.

These businesses are important employers, and many sit in industrial areas where software has barely arrived beyond accounting.

What already exists and the gap

• MRPeasy is a cloud ERP for small manufacturers with production planning, inventory and Gantt scheduling.
• Katana focuses on real-time inventory and production tracking with a visual planning board.
• ProShop ERP is a paperless ERP, MES and quality system for machine shops and contract manufacturers.

These are strong products, but they are priced in dollars per user, designed around Western accounting and inventory practice, and expect a planner who configures BOMs and routings carefully. A 20-person workshop needs something narrower: job travellers, a shared kiosk that workers with little computer experience can use by scanning, and a clear answer to "is this job late and is it profitable?"

What you build

Roles: owner or planner (orders, quotes, schedule, reports), supervisor (assigns work, records scrap), worker (scans at the kiosk, no login needed beyond a badge).

Flow: a customer order becomes a work order with a routing copied from the product template. A traveller sheet is printed with a QR code and moves with the physical job. At each work centre, a worker scans the traveller and their badge to start or finish. The WIP board and reports update live.

The technically interesting part

Two problems make this more than CRUD.

Finite capacity scheduling. Given work centres with daily hours, jobs with due dates and operations with standard times, propose a schedule. Start with a simple forward-scheduling heuristic (earliest due date first, each operation after its predecessor, no work centre over capacity) and measure it. Then try one improvement, such as prioritising by slack. This is a small version of the operations research problem factories spend fortunes on, and it gives you something to compare numerically.

Trustworthy time capture from a shared device. Workers forget to press finish. Two workers scan the same job. A kiosk reboots mid-shift. Design the event model (start, pause, resume, finish) with validation rules, auto-close of forgotten operations at shift end with a flag for supervisor review, and an audit trail. Actual cost reports are only as good as this data.

How to evaluate it

• Scheduling quality: on a dataset of at least 50 realistic jobs, compare your heuristic against a naive first-come-first-served baseline on late jobs and average lateness.
• Kiosk usability: time 5 workers doing start and finish scans; target under 10 seconds per action with no training beyond one demo.
• Data quality: in a pilot week, count operations auto-closed or corrected by the supervisor, and track whether it drops by week two.
• Business value: for 10 completed jobs, show quoted versus actual hours and discuss where estimates were off.

Suggested plan

1. Weeks 1 to 3: visit a workshop, photograph their job cards, list work centres and typical routings.
2. Weeks 4 to 8: products, routings, orders, work orders and printed travellers.
3. Weeks 9 to 12: kiosk scanning, event model, WIP board.
4. Semester 2, weeks 1 to 5: capacity scheduling heuristic and its evaluation.
5. Weeks 6 to 9: quote versus actual and scrap reports.
6. Weeks 10 to 14: one to two week pilot and final report.

Path to a product

Your first customer is a workshop in an industrial estate near you, reached through family, a faculty contact or a small industry association. Charge a monthly fee per site with unlimited kiosk workers, because per-seat pricing kills adoption on the floor. Grow by sector templates (furniture, fabrication, printing) with ready-made routings.

Scope warning

Do not build accounting, payroll, purchasing, or full inventory with stock valuation. Do not attempt an optimal scheduler; a measured heuristic is enough. Do not connect to machines or sensors; that is a different project.

Learn what you need on Sefism

• Backend roadmap : APIs, data modelling and background jobs for the scheduler.
• DSA cheat sheet : priority queues and greedy strategies behind the scheduling heuristic.
• Django REST Framework : a productive choice for the API if your team knows Python.
• PostgreSQL Documentation : constraints and transactions for trustworthy scan events.
• Refactoring.Guru: Design Patterns : state pattern for the operation lifecycle.

Get help with this idea

The best version of this project starts on a real shop floor, not in Figma. Tauseef can help you tighten the scope, choose a stack your team can finish with, and design an evaluation your examiners will respect. If you are new here, create a free account first.

• Discuss this idea with a mentor
• Send a query to Tauseef
• Book a 1:1 session