Dewel - Cold Chain Monitoring with Excursion Alerts and QA Release
Cold Chain Platform for an Ethiopian Logistics Provider
Designed and built a clickable cold chain monitoring platform for an Addis Ababa logistics provider that stores and moves vaccines, biologics and fresh fish: live temperatures from cold rooms, reefer trucks and box loggers, excursions measured against each product's profile, escalating alerts, and a QA disposition signed electronically.
Most of our work is under NDA. Per agreement, we can share the work but not the company details - all names and identifying information shown here have been changed.
Impact
Live dashboard, excursions list, excursion detail, QA review queue and the QA review and disposition workspace, with a critical alarm strip and Acknowledge button on every page.
Refrigerated 2-8 °C, frozen -25 to -15 °C, controlled room 15-25 °C, fresh fish 0-4 °C on ice and frozen fish -18 °C or colder, each with its stability budget, source and first actions.
Release, quarantine or reject is recorded per lot and signed with a password re-entry. A critical excursion then goes to the QA manager for a second signature.
13 QA reviews, 8 cold rooms at 2 sites, 14 reefer trucks on trip, 15 shipments in transit, 9 customers and 12 products, from measles-rubella vaccine to frozen Nile perch.
The Problem
Gullele Coldline ran its cold chain on three disconnected systems: a cold room alarm panel that beeped in an empty control room at night, the reefer vendor's telematics site that only the fleet manager used, and USB loggers downloaded days after delivery. Excursions were found after the product had been delivered, answering a pharmacist's "was this kept at 2 to 8 °C?" took QA half a day, and overdue calibrations and a missing mapping study cost the company in audits and tenders.
Our Solution
Dewel pulls readings from every cold room sensor, reefer unit and box logger, checks them against the product profile of whatever is inside, and opens an excursion with its duration, peak, degree-minutes, MKT and stability budget used. Alerts escalate until someone acknowledges and records a corrective action, and every excursion ends with a signed QA disposition per lot.
The Challenge: Proving Every Degree
Gullele Coldline stores and moves medicines and fish from 2 cold stores, 80 reefer trucks and a pool of insulated boxes. Each product has its own range and tolerance, and the record has to stand up to a good distribution practice inspection as well as a customer's quality agreement.
Alarms Nobody Answered
An alarm at 02:00 is only useful if someone acts on it. Drivers and warehouse staff work across regional roads with patchy signal, so alerts have to escalate by SMS and voice call, in Amharic or English, until a named person acknowledges.
One Rule Does Not Fit Every Product
A vaccine that must never freeze, a frozen biologic and fresh tilapia on ice each have different limits and a different allowed time out of range from the manufacturer's stability data. The same 9 °C reading can be minor for one and critical for another.
Decisions That Must Be Signed
Only a QA person may decide that product is fit for use. Customers ask for Part 11 style controls: unique users, re-authentication at signing, the meaning of each signature and an append-only trail of who changed what and why.
Records an Inspector Will Ask For
Good distribution practice expects calibrated sensors traceable to an accredited laboratory, temperature-mapped rooms and trucks, documented excursion handling and a temperature history per shipment. These lived in a binder and a set of screenshots.
Our Approach: From Cold Chain Rules to a Clickable Platform
Spec and Rules Research
Mapped 11 user roles from control room operator to the QA manager as Responsible Person, 43 screens across the web console, a driver and warehouse mobile app and a public certificate check, and the data model for assets, sensors, loggers, readings, product profiles, alarms, excursions, QA reviews, signatures, calibrations and mapping studies. Wrote down the rules the design follows: EFDA and WHO good storage and distribution practice, Part 11 and Annex 11 style signatures, MKT, Codex for fish and Ethiopia's Personal Data Protection Proclamation.
Clickable Demo on a Production UI Kit
Built 5 key screens on a production dashboard kit: the live dashboard with feed health, cold room tiles, fleet grid, shipments in transit and open alarms, the excursions list with bulk assign, the excursion detail with its temperature trace, escalation timeline, corrective actions and root cause, the QA review queue, and the review workspace with evidence tabs and per-lot disposition. All actions run on local state.
What a Production Build Adds
Integration with reefer telematics (such as Thermo King TracKing or Carrier Lynx Fleet), Teltonika trackers, cellular and USB loggers, Vaisala or LoRaWAN room sensors, ethio telecom SMS and Twilio voice, and the ERP. A time-series store, the escalation engine, staff auth by role, an append-only audit trail, and the remaining screens: live map, shipments and certificates, asset registers, calibration, mapping, reports, settings and the mobile app.
Why It Works: The Product Profile Decides
Problems First, Everywhere
A critical alarm runs as a red strip under the top bar on every page with an Acknowledge button. The fleet grid sorts problem trucks first, parked trucks collapse into one row, and late feeds and offline devices are named instead of hidden.
Every Figure QA Needs in One Place
The excursion detail shows the trace from 2 hours before to 2 hours after with the range shaded, limits dashed and door-open periods as bars, then peak, time above limit, degree-minutes, MKT and budget used, such as 38 of 120 minutes above 8 °C.
Nothing Skips a Step
Send to QA stays disabled until the excursion has ended, a corrective action exists and the root cause is filled in. Every alert sent, failed SMS and missed call sits on the escalation timeline with who acknowledged and how.
A Person Signs, Not the System
Dewel counts down the stability budget but never releases product itself. The QA officer weighs the evidence, including whether each sensor was in calibration on the day, and signs release, quarantine or reject per lot with the meaning of the signature shown.
Related case studies
View all: Logistics & supply chain
Dunnage - Shipment Tracking Portal and Ops Console for a Freight Forwarder
Designed a shipment tracking portal and operations console for an Antwerp freight forwarder moving ocean and air freight between Asia and Europe, and built the ops console as a clickable demo: shipment files with predicted and carrier ETAs, milestone timelines that show their sources, an exception queue against response targets and carrier data feed health.
Recua - Dispatch and Route Optimisation Console for a Regional Distributor
Designed and built a clickable dispatch console for a Guadalajara distributor of building materials and beverages: import tomorrow's orders, plan routes for four depots on a map, route list and timeline, check every manual change against the fleet's limits, publish before the 20:00 deadline and watch the day live.
Enterprise Software Engineering - Human + AI
Contact Us
© 2026 - Copyright. All Rights reserved Oceanlab
Location
Global Headquarters
21 Merchant Road, 3rd Floor, Singapore 058267
Main corporate office and leadership operations
North America Office
548 Market Street, PMB 12345, San Francisco, CA 94104
Client operations and business development
India Office
402 Shivalik Shilp II, SG Highway, Ahmedabad 380015
Engineering and delivery center
Enterprise Software Engineering - Human + AI
Contact Us
Locations
Global Headquarters
21 Merchant Road, 3rd Floor, Singapore 058267
Main corporate office and leadership operations
North America Office
548 Market Street, PMB 12345, San Francisco, CA 94104
Client operations and business development
India Office
402 Shivalik Shilp II, SG Highway, Ahmedabad 380015
Engineering and delivery center
© 2026 - Copyright. All Rights reserved Oceanlab