About

Starting Semester: Fall 2026
Assigned: No
Location: Grand Rapids

Steelcase

Client Profile

Steelcase is a global, publicly traded company leading our industry with fiscal 2025 revenue of approximately $3.2 billion and nearly 10,000 employees around the world. Steelcase has its global headquarters in Grand Rapids, Michigan. Steelcase brands offer a comprehensive portfolio of workplace products, furnishings and services, inspired by over 100 years of insight gained serving the world’s leading organizations. https://www.linkedin.com/company/steelcase/

Project Description

2. Industry Sponsor Overview
Steelcase supports dealers by configuring and quoting custom modifications to standard products. When a quote becomes an order, downstream groups—including design engineering, product data, and manufacturing—execute the customized build. The quoting team handles 10,000–15,000 quote requests annually, spanning diverse product families and requiring specialized product expertise. Due to increasing dealer expectations for rapid responses, and the difficulty of locating similar prior quotes, the system experiences inefficiencies and rework.

3. Problem Background & Motivation
The current quoting environment faces several systemic challenges:
• High transaction volume strains existing capacity.
• Dealers expect turnaround times faster than current processes allow.
• Past quotes are difficult to search, resulting in duplicated effort.
• Product‑specific expertise is required to determine feasible modifications.
• Repeated quoting of similar or identical requests wastes time and contributes to long cycle times.
Historical quoting data is a valuable but underutilized asset. Applying industrial engineering, machine learning, information retrieval, and operations optimization could significantly reduce quoting lead times and improve dealer satisfaction.


4. Problem Statement
How can the sponsor leverage historical quote data to identify duplicate quotes more efficiently, automate portions of the quoting process, and improve quote turnaround time while maintaining accuracy? The goal is to design a system that reduces manual searching workload, increases reuse of past solutions, and provides decision support for incoming quote requests.
5. Project Objectives
Analytical & Systems Objectives:
• Analyze historical quoting data to identify patterns, clusters, and similarity metrics.
• Develop a predictive model that estimates the likelihood an incoming request has a near‑duplicate.
• Engineer a text‑ and attribute‑based search tool optimized for large‑scale quote retrieval.
Operations & Human‑Systems Objectives:
• Reduce quoting cycle time via faster identification of prior work.
• Build a decision‑support interface for quote representatives.
• Improve consistency, accuracy, and throughput.
Automation & Optimization Objectives:
• Evaluate AI, predictive models, and ML methods.
• Prototype preliminary quote recommendation system.
• Provide confidence scoring and quantify operational benefits.

6. Proposed Deliverables
1. Quote Retrieval & Search Optimization Tool
2. Duplicate‑Quote Probability Model
3. Automated Preliminary Quoting Engine
4. Detailed Technical Report
5. System Architecture & Deployment Roadmap
6. Measurable Impact Assessment

7. Expected Impact & Value for Sponsor
• 30–50% reduction in quote turnaround time
• Reduced rework via duplicate detection
• Improved dealer satisfaction
• Scalable quote automation
• Higher quoting consistency

Skills

• Data analysis and statistical modeling
• Machine learning classification and clustering
• Natural‑language processing
• Operations engineering and workflow analysis
• Human‑factors & decision‑support system design
• Simulation or queuing analysis (optional)
• Optimization and systems integration planning

Data Access Requirement