Static rules within legacy rate shopping software were not optimized for maximum optimization due to the high variability of shipment costs and speed for a major portion of annual volume.
The older nature of deployed technology meant requested operational changes took months. Long feedback loops and reliance on external IT teams constrained innovation.
Combined with the lack of visibility into delivery performance, experimentation in service of optimization became all but impossible.
In a classic chicken-or-the-egg situation, the company needed a way to understand impact on operational performance yet were dependent upon expensive and major technology changes for those insights.
But to justify the investment to make technology changes, they needed performance results first.
The Shipium Solution
Shipium's Simulation offering unstuck their Catch-22 by providing performance insights without the requirement for technology changes. The company was able to run historical shipments through the Simulation offering, and tweak various customizations to understand different results, like how cost-per-package would change if including an additional regional carrier's rates, or if volume shifts would impact carrier discounts.
Example Optimization Simulations
Base Rate Optimization: Taking existing rate cards and simulate historical shipments using different selection criteria, for example different rules to filter out certain carriers to certain zones, to see how cost structures changed.
Accessorial Avoidance: Simulate the impact on volume shifts and costs if certain accessorial and surcharges were changed.
Transit Optimization: Impose a date constraint on historical shipments (e.g. % must be delivered in 2-days), then see volume and performance impact, including visibility into smart downgrades and required upgrades, using Shipium's machine learning-based transit data model.
Discount Optimization: How volume would shift and cost structures would be impacted if carrier volume discount tiers changed.
Example Planning Simulations
Carrier Expansion: Simulate historical shipping data with new carriers and rates added as options for selection to evaluate how volumes would shift and performance would be impacted.
Network Expansion: Simulate historical shipping data with new fulfillment origins (i.e. warehouses or additional 3PL nodes) added, then make assumptions based on volume shifts and performance impact based on which shipments would have come from that origin based on closest proximity.
Store Expansion: With an existing store network in place, simulate certain stores becoming shipping options. Map carrier and service methods and their rates to stores, and make an assumption on % of shipments that would originate from the store network, and view volume and performance impact.
The company turned to Shipium to simulate their existing fulfillment network and carrier mix with a goal of optimizing base rates and accessorial avoidance while still maintaining existing carrier discount tiers. They then layered in transit optimizations by imposing a desired delivery date using Shipium's dynamic TNT model along with Business Days of Transit (BDOT) 2 and 3 models.
- 42% of annual shipments would have been processed differently with Shipium due to human-driven rules in the legacy system missing situations where a more optimized carrier service method was the better fit per the selection criteria. Of the 42%, Shipium reduced CPP by $0.89.
- 19% of annual shipments produced a different Time-in-Transit value per Shipium's dynamic model versus carrier defaults, resulting in a shift of smart downgrades applied to cheaper service methods.
- Total CPP reduced $0.36, resulting in a simulated savings around $29.6 million across 80M annual domestic parcel shipments.