Currently, asset tracking requires manual effort by going to rail interfaces and submitting queries. [REDACTED] is not alerted to changes. Railcar return prediction is based on rule of thumb calculations impacting allocation, rather than being driven by historical data. In terms of optimizing allocation within railcars, the current workflow uses spreadsheets that require manual input from multiple data sources to budget customer orders. The experience on the customer side is fragmented, as well. Customers are informed about order problems, demurrage and invoices manually, which leads to varying customer experience.
Artificial intelligence techniques have a great capacity for identifying links and connections between disparate data elements. Over time, a system can evolve towards greater speed and accuracy in making such associations, whether causative or not. By building predictive systems that leverage historical data, [REDACTED] can optimize their entire organization for efficiency while never losing sight of the necessary concerns for the safety of human workers, especially those working in proximity to dangerous chemicals.
Hypergiant partnered closely with [REDACTED] on a product definition and proof of concept demo engagement, which we began by working with client stakeholders to identify and document the foundational architecture, needs, preferences, requirements, feature sets/functionality, user experiences, user interfaces, KPIs, constraints and other critical elements needed to inform the platform’s development and prepare [REDACTED]’s IT and data environment for successful integration and deployment. There are three key feature capabilities our solution addresses, including:
Railcar Tracking & Prediction - Using APIs, [REDACTED] can have near-real time tracking for their entire fleet. The system can monitor status changes and send timely alerts. As the fleet grows, the ability to track railcars using APIs will maintain the fleet and serve to only alert when people need to take action, thereby saving time.
Optimal Allocation - Hypergiant uses Constraint Programming to allocate contracts and confirmed orders into an optimal delivery plan. Sulphuric acid inventory, railcar availability and other constraints are balanced with requested shipment dates to optimize customer satisfaction and asset utilization. This will help [REDACTED] scale by efficiently accommodating additional orders and customers.
Customer Experience - [REDACTED] Fleet Manager can be configured to trigger timely alerts based on customer needs, impacting events across the system. Customer-facing dashboards can also be used to confirm and track orders. Customer Alerts help group alerts from different parts of the system making it easier to manage customer notifications with minimal effort. Customizable customer dashboards allow 360-degree view of future and shipped orders.