Operations Analytics Forecasting Excel 2025

Operations &
Forecasting Model

A multivariate regression engine built on 18 months of historical sales data, paired with an interactive KPI dashboard that translates demand signals into inventory decisions — reducing carrying costs and improving delivery performance.

94%
Forecast
accuracy
35%
Carrying cost
reduction
28%
Delivery
improvement
15+
KPIs
tracked
01 — Overview

The problem with gut-feel inventory

Most small-to-mid operations teams make inventory and staffing decisions reactively — responding to stockouts or backlogs after they've already cost revenue or customer trust. Without a reliable demand signal, even experienced operators overbuy in slow periods and understock when demand spikes.

This project set out to replace that guesswork with a repeatable, data-driven forecasting workflow. The goal: build a model accurate enough to inform proactive inventory positioning, and surface the right operational KPIs so teams can catch issues before they compound.

The result is a two-part Excel system — a validated regression model for demand forecasting, and a live KPI dashboard that turns the model's output into actionable operational metrics.

Project type
Solo — end-to-end analytics build
Tools used
Excel, Solver, VBA, Pivot Tables
Data
18 months historical sales records
Validation method
Holdout test set — 20% split
Deliverables
Regression model + interactive KPI dashboard
Year
2025
02 — Problem framing

Two challenges,
one integrated solution


Challenge 1 — Demand signal

Without a structured forecasting method, inventory decisions were driven by trailing averages and intuition. Lead times varied, and there was no way to anticipate the demand spikes that drove the largest carrying cost overruns. The team needed a quantitative demand model that could generalize across seasonal variation.

Challenge 2 — Operational visibility

Even with a better demand estimate, operators had no unified view of how the operation was performing day-to-day. Machine utilization, inventory turnover, and order fill rates all lived in separate reports. There was no single dashboard surfacing leading indicators early enough to act on them.

03 — Methodology

How the model
was built


01

Data collection & cleaning

Compiled 18 months of historical sales records across product lines. Cleaned for duplicate entries, corrected date inconsistencies, and flagged outliers using IQR-based thresholding. Transformed raw transaction logs into a structured weekly time series ready for regression.

Excel Power Query Outlier detection Time series formatting
02

Feature engineering

Constructed predictive features from the cleaned dataset: rolling 4-week moving averages, month-of-year seasonality indicators, lagged demand values (t-1, t-2, t-4), and a trend variable capturing longer-term growth. Evaluated correlations with a heatmap before including features in the final model.

Lagged variables Seasonality encoding Correlation analysis
03

Multivariate regression model

Built the regression model using Excel's Data Analysis Toolpak with the engineered feature set. Evaluated R², adjusted R², F-statistic, and p-values for each predictor. Iteratively removed low-significance features and refit until achieving a model with strong explanatory power and minimal multicollinearity.

Multiple linear regression R² = 0.94 Adjusted R² p-value filtering
04

Holdout validation

Reserved the final 20% of the time series as a holdout test set — not used during model training. Ran the fitted model against this held-out period and measured mean absolute percentage error (MAPE) and root mean square error (RMSE). Validated that the model generalizes beyond the training window before use in production decisions.

Train/test split MAPE RMSE Generalization check
05

KPI dashboard design

Designed an interactive Excel dashboard consolidating 15+ operational KPIs into a single view. Core metrics include machine utilization rate, inventory turnover, order fill rate, average lead time, and backlog volume. Used dynamic named ranges and slicers for filtering by product line and time window. Color-coded thresholds flag metrics approaching risk territory.

Excel dashboards Dynamic charts Conditional formatting Slicers Named ranges
06

Inventory positioning logic

Translated model forecasts into reorder point calculations using a safety stock formula that accounts for demand variability and lead time uncertainty. Built a decision layer in the dashboard that outputs recommended order quantities per SKU based on current stock levels and the upcoming forecast horizon.

Reorder point calc Safety stock formula Excel Solver Decision logic
04 — Results

What the model
delivered


94%
Forecast accuracy
Validated on holdout test set
−35%
Carrying cost reduction
Simulated vs. baseline inventory policy
+28%
Delivery performance
On-time order fulfillment rate
Forecast accuracy (holdout) 94%
Carrying cost reduction 35%
Delivery performance improvement 28%
KPIs tracked in dashboard 15+
Historical data used (months) 18 mo


Proactive inventory positioning replaced reactive restocking

The regression model gave the team a 3–4 week forward demand signal, allowing reorder decisions to be made before stockouts occurred — eliminating the emergency procurement costs that drove most carrying cost overruns.

Machine utilization surfaced as a leading indicator

The dashboard revealed that utilization rate at key production stages consistently preceded backlog buildups by 10–14 days. Teams could now intervene (staffing, scheduling) before queues formed.

Seasonality encoding improved peak-period accuracy

Adding month-of-year binary indicators reduced MAPE by ~8 percentage points vs. a model without seasonality features — the single highest-impact feature engineering decision.

Safety stock formula reduced stockout frequency

The reorder point calculation, incorporating demand standard deviation and lead time variance, cut stockout events in the simulated period by over 60% versus a simple moving average policy.

05 — Technical details

Under the
hood


Model specification

Ordinary least squares via Excel Data Analysis Toolpak
Dependent variable: weekly unit demand
Predictors: 4-week MA, lagged t-1/t-2/t-4, trend index, 12 month dummies
80/20 train-test split — no data leakage
Adjusted R² = 0.91, F-stat significant at p < 0.001

Dashboard architecture

Dynamic named ranges bound to slicer selections
Conditional formatting thresholds at 80%, 90%, 100% utilization
OFFSET/INDIRECT formulas for rolling window calculations
Solver-based reorder quantity optimization
VBA macro for one-click data refresh and export

KPIs tracked

Machine utilization rate (by station)
Inventory turnover ratio
Order fill rate & on-time delivery %
Average cycle time & lead time variance
Backlog volume by queue stage
Carrying cost per unit & total holding cost

Limitations & assumptions

Linear model assumes demand relationships are stable over time
Does not account for supply-side disruptions or price elasticity
Reorder logic assumes constant lead time (future: stochastic)
Holdout period may not fully reflect future regime changes
Microsoft Excel Data Analysis Toolpak Excel Solver VBA / Macros Pivot Tables Power Query Conditional Formatting Dynamic Charts Named Ranges Multivariate Regression
06 — Reflections

What I'd do
differently


Lesson learned

Forecast early, validate often

The biggest operational errors came from waiting too long to act on demand signals. A model that's 85% accurate and acted on early outperforms a 94% model used reactively. Earliness of intervention matters as much as model precision.

What I'd improve

Move to a rolling retrain

The static model was trained once. A production system should retrain on a rolling window — incorporating the last N weeks of actuals — so the model adapts to demand shifts rather than drifting out of calibration over time.

Next iteration

Python + automated pipeline

Excel is powerful but brittle at scale. The next version would move the regression and KPI logic into Python (pandas, statsmodels, Streamlit), with an automated pipeline that pulls live data, retrains the model, and pushes updated forecasts to a web dashboard.

Broader takeaway

Bottlenecks are data problems

The stations with highest queue buildup were the ones with least visibility — not because they were inherently slower, but because there was no early warning. Better instrumentation (more KPIs, lower latency) is often worth more than additional capacity alone.