Weight-Sensing Stock Bins for Pharmacy and Spare Parts Shelves
A final year project that builds load cell stock bins for pharmacies, spare parts shops and workshop stores: each bin converts weight to item count, flags low stock and logs every pick. A small web app turns the readings into reorder lists, so owners stop running out of fast movers and stop counting shelves by hand.
Intermediate
A small team, or one strong student willing to learn something new.2 semesters · 2 to 3 students
7 to ship
3 optional extrasSuggested stack
What you should ship
- A modular bin base with a load cell and amplifier; several bins share one ESP32 controller per shelf.
- Per-item unit weight learning: the user places a known count once and the system stores unit weight and tolerance.
- Count calculation with a confidence flag when the weight does not match a whole number of items within tolerance.
- A web app listing items, current counts, reorder levels and a generated reorder list grouped by supplier.
- A pick log that records each removal and restock with time and quantity, and flags removals outside business hours.
- A shelf-level LED that lights for low stock or for the bin a staff member searched for in the app.
- A two-week pilot on one real shelf with a comparison of system counts against manual counts at the end of each day.
If you have time left
- Import sales from a point-of-sale CSV and reconcile them against weight-based picks to spot shrinkage.
- Predict days until stock-out per item from pick history.
- Battery-powered bins that report over BLE to a shelf gateway.
The problem small stores have with counting stock
A pharmacy, an auto spare parts shop or a factory maintenance store holds hundreds of small, fast-moving items: strips of common medicines, bolts, fuses, bearings, filters. Stock is counted by hand, usually only when something runs out. The owner discovers a stock-out when a customer asks for it, and discovers theft only at a yearly count. Barcode-based inventory software helps only if every sale is scanned, which is rarely true in a busy small shop.
What already exists and the gap
- eTurns TrackStock SensorBins, powered by Mettler-Toledo load cells, convert bin weight to quantity and trigger reorders; aimed at distributors running vendor-managed inventory for industrial customers.
- Bossard SmartBin does the same for fasteners and C-parts in factories as part of Bossard's supply service.
- Point-of-sale inventory apps track stock from sales scans; accurate only when every movement is scanned.
The gap is a bin system that a single shop or small workshop store can buy outright, that runs on hardware costing a small fraction of industrial systems, and that is honest about when its count is uncertain.
What you build
- Bins: shelf-mounted plates with a load cell each; a shelf controller reads several bins.
- Firmware: calibration, filtering, change detection and publishing counts and events.
- Backend: item catalogue, unit weights, reorder levels, pick log and reorder list generation.
- Web app: inventory table, low stock view, reorder list and pick history.
Roles: the owner, who sets up items and reorder levels, and staff, who search for items and restock.
The technically interesting part
Converting weight into a count is harder than dividing by unit weight. Unit weights vary (medicine strips with different blister packs, bolts from different batches), load cells drift with temperature and creep under constant load, and people lean on shelves. You need a tolerance model: for each item, estimate the spread of unit weight and decide when a reading is a confident count and when it is ambiguous. You also need event detection: when weight changes and settles, record a pick or restock with a quantity. Show evaluators the accuracy by item type, how often the system flagged uncertainty, and how drift correction behaves over days.
How to evaluate it
- Count accuracy for at least ten item types with different unit weights, reported as percentage of exact counts.
- Drift over 48 hours with constant load, before and after your compensation.
- Pick detection: precision and recall of pick events against a video or manual log for a trial day.
- Cost per bin compared with the item value it protects.
Suggested plan
- Month 1: shop visit, pick 20 items, weigh samples to measure unit weight spread.
- Month 2: bin hardware, calibration and filtering.
- Month 3: event detection and count confidence model.
- Month 4: backend, reorder logic and web app.
- Month 5: shelf LEDs, pick log and alerts.
- Month 6 to 7: shop pilot, accuracy study and report.
Path to a product
The first customer is a pharmacy or spare parts shop owner who loses sales to stock-outs of a few fast movers. Start with one shelf of the top 20 items, sold as a kit with a monthly software fee. Distributors that supply these shops could fund bins to get automatic reorders, the same model industrial suppliers use. Factory maintenance stores are a second market.
Scope warning
Do not build a full point-of-sale, accounting or supplier ordering system. Do not cover every item in the shop; choose fast movers where weight works well. Loose items of very different weights in one bin will not work, so say so.
Learn what you need on Sefism
- Backend roadmap : designing the catalogue, events and reorder logic.
- Fullstack cheat sheet : connecting the web app, API and database cleanly.
- Django REST framework documentation : a quick way to build the inventory API.
- PostgreSQL Documentation : modelling items, bins and pick events.
- The Mom Test : asking shop owners about stock-outs without leading them.
Get help with this idea
The count confidence model is the piece most worth discussing before you build. If you want a second opinion on scope, hardware choices or how to set up the evaluation before you commit, bring your plan and your bill of materials and we will pressure-test it together. New here? Create a free account first.
Ideas and guidance, not finished projects
These are project ideas and scoping guidance, published free for students to use as a starting point. I do not build, write, or sell final-year projects, and I do not complete coursework for anyone. Take an idea, make it yours, and build it.
Not sure this one fits you?
I help students pick an idea that matches their skills and their deadline, then scope it down until it is finishable. Guidance only, never done for you.
Stuck on something specific?
Writing only gets you so far. If you want an answer to your situation rather than the general case, book a session and we will work through it together. Sessions are free for approved Sefism members, and a few slots open each week.
Follow along
New writing, resources and project ideas land here first.