Flexee Analytics
Students run the supply chain of a residential power manufacturer — and the decisions each quarter come straight from the analysis they just did. Download real data, choose your technique, make the call, and watch the simulation reveal whether the analysis was sound. The grade is in the results, not just the report. Built for analytics-focused business and data courses.
Six analytics modules. Real data packages. Direct validation loop.Built on Flexee Supply Chain
Same simulation engine, same financials, same credit scoring, same competitive teams. What changes in the Analytics Edition is the focus: the operational complexity drops away and students spend their time on the analysis that drives the decisions. Everything on this page is what the Analytics Edition adds.
Six analytical methods, each taught the way they're used in practice. Time-series forecasting. Total Cost of Ownership. RFM segmentation. Location optimization. Conjoint analysis. Weighted scoring. Real techniques, real data, real consequences when the analysis is wrong.
Fifty-two weeks of regional sales history with seasonality, trend, and noise. Students choose between moving averages, exponential smoothing, and regression — fit, validate, and submit a quarterly forecast. MAPE shows whose model beat intuition and whose intuition beat the model.
Time-series forecastingCost, quality, lead time, and risk data across six global and regional suppliers. Students build a TCO model — landed unit cost plus quality cost plus disruption risk premium — and select primary and backup suppliers. The cheapest supplier almost never wins.
TCO analysisTwo hundred customer records with purchase recency, frequency, and monetary value. Students score and segment into Champions, Growth, At-Risk, and Other — then allocate marketing spend across the four segments. Bad segmentation produces visible churn next quarter.
RFM segmentationSixteen demand locations and ten candidate DC sites with fixed costs, variable costs, and distance data. Students set up a facility location problem, evaluate single-DC and multi-DC scenarios, and commit. Perfect Order on-time tells them whether the optimization was sound.
Location optimizationConjoint survey data from twenty-five respondents rating eight product configurations. Students decompose preference into part-worth utilities, identify the configuration with the highest predicted share, and price it. The market either confirms the analysis or it doesn't — and four quarters of P3 sales tell the truth.
Conjoint analysisCountry profiles for Canada, EU/Germany, and APAC/Japan — market size, entry cost, regulatory risk, and strategic fit. Students build a weighted scoring model with their own assumptions about what matters most, rank the three options, and choose. Expansion revenue plays out over six quarters and the assumptions get tested.
Weighted scoringSix features that turn a simulation into an analytics course where the analysis actually matters — because the simulation grades it.
Analysis leads to decision. Decision leads to result. Result tells students whether the analysis was sound. The simulation becomes the feedback mechanism for every analytical technique — not a grading rubric written by someone who already knows the answer.
A dedicated screen where students access their data packages, track forecast accuracy over time, and model what-if scenarios before committing. Six data package cards, one per module. Download, analyze, decide, submit.
Each quarter, students declare whether their forecast came from intuition (GUT) or analysis (MODEL). The system persists both. At simulation's end, students see their own MAPE for each — and the comparison is usually the most memorable moment in the debrief.
Data packages are CSV files. Students work in whatever they already use — Excel, Python, R, Tableau, SAS. No proprietary black box. The analysis happens in the tools students will use in practice, and the output is what they submit.
The Corporate Edition asks for twenty to thirty operational decisions per quarter. Analytics Edition strips that back to six — one per module. The complexity is in the analysis, not the operations. Students can focus their effort where the learning is.
Most analytics assignments end when the report is submitted. Here, a bad forecasting model means overstocking or stockouts. A weak segmentation means at-risk customers leave. A careless scoring model means the wrong market entry. The grade shows up in the simulation — not in the margins of a PDF.
Configurable from 8 to 24 quarters, with 20 as the default. Students download data packages between quarters, do their analysis in their preferred tools, and submit decisions. Built for analytics courses in MBA programs, data science cohorts, and business analytics electives.
Students pull this quarter's data packages from the Analytics Workbench — CSV files they can open anywhere.
Teams work in their tools of choice — Excel, Python, R, Tableau — applying the analytical methods they're learning.
Decisions flow from the analysis. Students declare GUT or MODEL and submit through the simplified cockpit.
Results release. MAPE, Perfect Order, revenue, and market share reveal whether the analysis held up under real competition.
The central idea
"Most analytics assignments have no consequences. Students do the regression, submit the report, get a grade. Here, if their forecasting model was bad, they overstocked or understocked and their Perfect Order suffered. The grade is in the simulation results, not just the analysis document."
From the Flexee Analytics design brief
Your implementation guide
Sales, implementation, and training for Flexee Analytics. He'll help you map the simulation to your analytics syllabus — which modules fit your curriculum, how to pace the quarters, and how to integrate the analysis into existing assignments.
CPIM · CSCP · CLTD · CTSC
Book a thirty-minute walkthrough. We'll show you the Analytics Workbench, walk through a sample module, and help you think through where Flexee Analytics fits your course.