BUS 721 Week 2 Business Intelligence Example

Seeing Every Builder Order Across Three Plants: Business Intelligence for Delivery Performance at Tri-State Cabinet

[Student Name]

University of Phoenix

BUS/721: Issues in Optimizing Operations

Week 2 Assignment

[Instructor Name]

[Date]

Tri-State Cabinet Company and all data are composites written for a model paper.

Tri-State Cabinet Company, the composite cabinet maker with plants in Pennsylvania, Maryland and Virginia, has adopted a strategy of fast, reliable delivery to home builders. Week 1 found that it cannot see order status across plants: customer service staff phone each plant, plant schedulers keep spreadsheets and late-order reports are compiled by hand each month. An operation cannot manage what it cannot see, and a company that learns about late orders from angry customers is managing by surprise. This paper examines how business intelligence can support Tri-State's optimization.

Defining Business Intelligence and Analytics

Business intelligence refers to the technologies and practices for collecting, integrating and reporting data to support decisions. Analytics extends this to diagnosing causes, predicting outcomes and recommending actions. Together they turn transactional data into information managers can act on.

The Evolution of BI and Analytics

Chen et al. (2012) traced the evolution of business intelligence and analytics from a first stage focused on structured data in data warehouses and reporting, through a second stage incorporating web and unstructured data, to a third stage using mobile and sensor data. They emphasized that analytics capability depends on data, technology and skilled people.

Information Management Capability

Mithas et al. (2011) studied firms and found that information management capability, the ability to provide data and information to users with appropriate accuracy, timeliness and security, influenced firm performance through customer management, process management and performance management capabilities. The value of information comes through how it improves these management capabilities.

Data-Driven Decision Making

Brynjolfsson and McElheran (2016) found that data-driven decision making spread rapidly in U.S. manufacturing plants and that adopters had higher productivity. They noted that benefits depended on complementary investments in information technology, education and management practices.

Where Tri-State's Data Live

Order data sit in a legacy enterprise system at headquarters; production schedules in each plant's spreadsheets; quality data on paper rework tags; shipping data in a carrier's portal. Definitions differ: one plant counts an order as shipped when the first box leaves, another when the last does.

Assessing Data Maturity

On a simple maturity scale from ad hoc reporting to predictive analytics, Tri-State sits near the beginning: manual reports, inconsistent definitions, no single source of truth and few analytical skills outside finance.

Decisions BI Should Support

Key decisions include which orders to prioritize each day, when to warn a builder of a delay, where to add capacity, which defects to target and which suppliers cause delays. Each requires timely, accurate data.

The Order Visibility Dashboard

The first application is a dashboard showing every builder order's status across plants: components scheduled, produced, finished, packed and shipped, with flags for orders at risk of missing their promised date. Customer service, schedulers and plant managers will use the same view, ending the need to phone each plant for answers.

Diagnosing Late Orders

An analysis of six months of order data, assembled manually for this paper, found that 46 percent of late orders were delayed by finish rework, 27 percent by missing components from another plant, 15 percent by supplier hardware and 12 percent by scheduling errors. This points improvement efforts toward quality and cross-plant coordination, consistent with the sand cone sequence discussed in Week 1.

From Descriptive to Predictive

Once descriptive dashboards are reliable, Tri-State can predict which orders are likely to be late based on style mix, plant loads and supplier lead times, allowing earlier action. Predictive models should follow, not precede, good data and basic reporting.

What the Sample Revealed About the Data

Assembling six months of order data by hand exposed problems leaders had not known about. About 9 percent of orders lacked a recorded promise date, so their lateness could not be measured. Plant timestamps were entered at shift end rather than when work occurred. Two plants used different codes for the same door style. These findings show why the first phase of BI must focus on data capture and definitions rather than on visual dashboards, which would otherwise display confident-looking but unreliable numbers.

Integrating Systems

Integration will connect the order system, plant scheduling, scanning at each production stage and the carrier's shipping data into a single data store refreshed hourly. Integration is the most technical and expensive part of the project, and a phased approach, starting with the plant that handles the most builder orders, limits risk.

Choosing Measures

The dashboard will center on on-time complete delivery, defined as all cabinets in an order arriving by the promised date, along with lead time, orders at risk, rework rate by line and supplier delivery performance. A small set of shared measures is more useful than dozens of reports that each plant interprets differently.

Data Quality

Data quality problems include inconsistent definitions, missing timestamps and manual entry errors. A data cleanup project will standardize definitions of order status, require scanning of components at each stage and assign data owners in each plant.

Data Governance

A governance group of operations, IT and finance leaders will own data definitions, approve new reports and set access rules. Clear ownership prevents the conflicting numbers that undermine trust in dashboards.

Building Analytical Skills

Two analysts will be hired, and plant schedulers and supervisors will be trained to use dashboards and basic analysis. Skills, as Chen et al. noted, are as important as tools.

Adoption and Routines

Dashboards matter only if used. Daily production meetings at each plant will review at-risk orders from the dashboard, and weekly cross-plant calls will resolve component issues. Leaders will use dashboard data in reviews instead of manual reports.

The Human Side of Visibility

Visibility can feel threatening. Plant managers worry that a dashboard showing late orders by plant will be used to blame them. Leaders will present the dashboard as a tool for solving problems, review it in joint meetings rather than in individual performance reviews at first and celebrate improvements. Over time, as trust builds, plant measures can be added to scorecards.

Investment and Return

The BI investment, including data integration, scanning hardware, dashboard software and analysts, is about $1.4 million over two years. If on-time complete delivery rises from 73 to 90 percent, retained builder business alone could be worth several million dollars a year.

Limits of the Evidence

Research on BI and data-driven decision making comes largely from large firms and cross-sectional data, which limits causal claims and transfer to mid-size manufacturers. Tri-State's own before-and-after measures will provide local evidence.

Research Questions

The case suggests questions: how mid-size multi-plant manufacturers build information management capability with limited resources, and whether order visibility alone improves delivery or requires accompanying process changes.

Conclusion

Business intelligence can give Tri-State the visibility its delivery strategy requires. Research links information management capability and data-driven decisions to performance, provided complementary practices and skills exist. An order visibility dashboard, late-order diagnosis, data cleanup, governance, training and daily routines form a roadmap from manual reports toward predictive analytics.

References

Brynjolfsson, E., & McElheran, K. (2016). The rapid adoption of data-driven decision-making. American Economic Review, 106(5), 133-139. https://doi.org/10.1257/aer.p20161016

Chen, H., Chiang, R. H. L., & Storey, V. C. (2012). Business intelligence and analytics: From big data to big impact. MIS Quarterly, 36(4), 1165-1188. https://doi.org/10.2307/41703503

Mithas, S., Ramasubbu, N., & Sambamurthy, V. (2011). How information management capability influences firm performance. MIS Quarterly, 35(1), 237-256. https://doi.org/10.2307/23043496