Transforming Product and Service Classification with AICA's AI Solution: a Case Study

The challenge of efficiently classifying vast catalogues of products and services according to the UNSPSC is a costly and time consuming task. A leading industrial supplier faced this challenge head-on by partnering with AICA, a solution at the forefront of AI-driven classification.

This case study delves into how AICA’s innovative AI technology, coupled with our specialised team, significantly enhanced classification efficiency and reduced operational costs.

The Challenge

The supplier’s catalogue comprised 145,000 items, each requiring precise classification.

 Traditional manual methods estimated the process would take approximately 1209 days, considering an 8-hour workday and an average classification time of 4 minutes per item.

 This labour-intensive approach not only demanded significant time but would also incur high costs, estimated at $0.50 per minute for which would result in a total cost of $362 500 to complete the 145 000 items.

The AICA Advantage

AICA introduced its AI-powered solution specifically designed to address the inefficiencies of manual classification. Leveraging advanced algorithms and machine learning, our AI was tasked with classifying the supplier’s extensive catalogue, promising both speed and accuracy.

Implementation and Impact

Speed

AICA’s AI technology processed the entire catalogue, delivering a fully classified item sheet in 20 days – a fraction of the time initially projected for manual classification. Our AI’s ability to classify items accurately and efficiently transformed the project’s timeline.

Accuracy

With an accuracy level that achieved a confidence score above 90% for 78% of all items, AICA significantly reduced the volume of items requiring further review. This high accuracy rate meant that only 22% of items needed manual classification, substantially decreasing the workload.

Efficiency

The AI classification cost stood at $0.45 per item, totaling $65,250 for the entire catalogue and $79 750 for the quality control process. This results in a total project cost of $145 000, a stark contrast to the projected costs of traditional manual classification methods.

Specialised Team Contribution

AICA’s specialised team played a pivotal role in the project’s success. While the AI handled the bulk of the classification, our team efficiently managed the manual classification and quality control of items that the AI flagged for further review. This collaboration ensured that the entire classification process was not only swift but also maintained the highest standards of accuracy.

Comparative Analysis

Time Saved

With AICA, the need for manual classification was dramatically reduced to 20 days, focusing only on a select portion of items. This strategic approach allowed for a significant reduction in project duration compared to traditional methods.

Cost Comparison

AICA’s solution offered a comprehensive cost advantage. The combined cost of AI classification and the specialised team’s manual review was significantly lower than the estimated costs for a fully manual process, highlighting the financial benefits of integrating AI into operational workflows.

Conclusion

By harnessing the power of AI and leveraging a specialised team for quality control, AICA not only expedited the classification process but also achieved considerable cost savings.

 This case study underscores the potential of AI to redefine industry standards, offering a scalable, efficient, and cost-effective approach to data classification that sets a new benchmark for operational excellence.