Jet-Lagged Planner ✈️

Project Duration

5 Weeks

Role

Developer, UX Designer

Softwares

Figma, Python, Streamlit

Work Type

ML Web App

Product Highlight

Constantly juggling making appointments across time zones? Jet-Lagged Planner has your back! Built for a wide range of industries such as Manufacturing, Aviation, and Construction, this smart planner tackles bulk scheduling with timezone conversion, conflict prediction, and AI-aided rescheduling customised to your own historical data to keep you away from a 3 AM site visit in a heatwave! With intuitive weather visualisations, CSV uploads, and ML-powered efficiency, it turns jet-lagged chaos into smooth, automated scheduling. Less manual work leads to fewer mistakes. This is your gateway to a more optimised work operation across the globe.

Problem Area

The goal of this project is to design a web app that optimises and streamlines the process of bulk scheduling inspections for large organisations operating across multiple time zones, and to overcome the technical obstacles that the SC team encounters. The design needs to reduce human error, improve efficiency, and predict conflict resolution strategies by incorporating factors like weather conditions, historical data, operational hours, and personnel availability.

Background

During my product design internship at SafetyCulture, I collaborated with 20+ designers actively developing the company’s design system for its main product, the SC web platform, which helps businesses streamline workplace operations and inspections. As part of the Scheduling team—known for handling the platform’s most complex feature—I followed an end-to-end design process to address a critical customer pain point. From customer calls, I observed that the current workflow of inspection scheduling does not allow multinational organisations such as Amazon to assign more than 2 timezones, forcing managers to manually match time zones for hundreds of sites, which can be a tedious task and is prone to human errors. To tackle this, I came up with 3 iterations of new timezone assignment work flows to address this issue and tested two of them through an unmoderated preference test on UserTesting.com.

Wireframes of 3 new creation flows of timezone which I created during my internship

Preference Test for the two Clickable prototypes that got the most team votes

Technical Demo

The model that supports the backend of the app is trained based on randomised data that includes entries like Site Name, Location, Time Zone, Inspection Type, Preferred Time Slot, Inspector Availability, and Weather Condition. And this is to foreshadow the possibility of allowing users to upload their own CSV of historical schedules for bulk scheduling, and the model will train dynamically on this data after being uploaded by users(this process is demonstrated through a basic Stream lit web-app). The code uses a Random Forest Regression model to predict optimal scheduling times based on K-Means Clustering, Weather Condition, Preferred Time Start, and Inspector Time Start. It simplifies scheduling by grouping inspections based on type and weather which helps identify patterns and categorises tasks efficiently. And it also realised timezone mappings (e.g. America/New_York) and conversion between times.

ML Next Steps：

• Introduce APIs from companies that specialise in weather forecasting.

• Expand Data Sources to train the model for more accurate predictions.

• Integrate AI tools such as Google Gemini for better conflict explanation.

Hi - Fi Feature Highlights

Homepage

By entering the web app, users can view their current schedule in an easy list with information such as assigned personnel , location, duration of inspection, and key words for inspection. The app also comes with interactive features that allow users to comment and share the schedule modifications within their organisations.

Data Upload Flexibility:

The "Take a Survey"/“Upload a CSV File” feature provides a clear and user-friendly process for uploading bulk data. By clicking "Take a Survey", users can manually enter their historical scheduling data through a Google form which gets converted to an CSV file if they do not their data organised in the format required by the app.

Weather Data Visualisation

The app will give users visualisations to view results including weather conditions, personnel availability, and operation hours, which are relevant to inspection scheduling, with UI highlighting conflicts with warnings (e.g., "Temperature Spike: 104°F at 03:00") and provides smart rescheduling options (e.g., "04:30 AM (60°F)").

Final Prototype

Team Feedback

“Smart rescheduling based on weather is a huge plus, especially in industries where extreme conditions impact work.”

-- Design Team For Web Platform

“Will love to see more AI suggestions for scheduling conflicts beyond weather, such as worker fatigue or availability constraints.”

-- Classmate

Future Advancements

• Implementing machine learning in my design allowed me to experiment with how AI can optimise scheduling efficiency and predict conflict resolution with a more human-centered approach.

• To further build on this idea, it will be helpful to find a way to incorporate historical worker performance data to factor in employee availability, workload distribution, and shift patterns for even more intelligent scheduling automation.