The Future of Forecasting MI MIT SCM M Capst ston one Proje - PowerPoint PPT Presentation

The Future of Forecasting MI MIT SCM M Capst ston one Proje ject ct Evan Humphrey | Federico Laiño Advisor isor | Inma Borrella

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Agenda I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 2 1 2 3 4 Introduction Methodology Findings Conclusions 1. Company Background 1. Overview 1. Characterization of Demand 1. Takeaways 2. Motivation 2. Discovery 2. Current Forecasting Process 2. Future work 3. Objective and Scope 3. Diagnosis 3. Forecast Accuracy Analysis 4. Demand Sensing Approaches 4. Suggestions for Implementing Demand Sensing

Introduction 1

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Company Background I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 4 1960s 1970s 1990s 2000+ 1980s Company ny Foun unde ded Develope loped d Etafil ilcon on 1 Day Lenses Market Leader Acquired ed by J&J Chief Optometrist Created first low-cost, JJVC gains and Division was renamed Originally ‘Frontier develops new daily disposable lens. maintains leadership to ‘ Vistakon ’. Contact Lenses’ from material, Etafilcon, Expanded globally to in the contact lens Developed automated Buffalo, New York. that allowed Brazil, Japan, Singapore market. production system, Later moved to production of soft and UK. Changed name leading to the creation Jacksonville, Florida lenses. to JJVC. of the Acuvue brand.

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Company Background I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 5 Europe, Middle East and Africa 21% of revenue US and Canada 22% volume 38% of revenue Limerick, Ireland 32% volume Production EVC – London, UK DC HDC - Tokyo DC Japan Jacksonville, FL 21% of revenue Production + DC Key Insights 26% volume $3Bn Business 4Bn Lenses Asia Pacific 22.000 SKUs 15% of revenue LATAM 16% of volume 3% of revenue 1% of volume

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño “ We must constantly strive to reduce our cost in order to maintain reasonable prices. Customers' orders must be serviced promptly and accurately. - Lines 3 and 4 of the J&J Credo 6

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Motivation I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 7 Market Context Contact lens global leader by market share but faces competition from other large companies and disruptive entrants. Production Capacity Cost Efficiency Owns high-end manufacturing lines Driven to continuously improve that are at near-maximum utilization forecast accuracy and capitalize with expansion requiring considerable on lower inventory costs and CAPEX and time. higher service levels. Forecast Accuracy Wants to explore the potential of demand sensing as a means to improve forecast accuracy

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Objective and Scope I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 8 Objective Scope Region Analyze the current J&J Vision Care forecasting process and propose suggestions for improvement, paying special attention to Demand Sensing approaches. Continental United States Production Facility Jacksonville, FL Brand Acuvue Jacksonville, FL Brand Family Production + DC 1-Day Moist (1DM) 1-Day Moist for Astigmatism (1DM-A)

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Methodology 2

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Overview I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 10 Tasks Sep Oct Nov Dec Jan Feb Mar Apr May Jun Phas ase I Objective and Scope Discovery Phas ase II Diagnosis Demand Sensing Approaches Capstone Write-up

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Discovery I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 11 Litera rature ture Review ew Interview erviews 01 02 SCM History Forecasting Forecasting Techniques Demand Planning Forecasting Measures 02 S&OP 01 Demand Sensing Case Studies Discovery Site Visit it 03 Jacksonville, FL 03 Production Facility Distribution Center S&OP Interviews

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Diagnosis I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 12 1 2 3 Demand Forecasting Forecast Characterization Process Mapping Accuracy Analysis Pareto Analysis Cycle Time Pareto Breakdown Time Series Framework Comparison with Distribution Data Inputs Alternative Statistics Forecasts Forecasts

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Demand Sensing Approaches I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 13 Latenc tency Reduct ction 01 Reduce cycle time between forecasts to take advantage of latest demand information updates. Meas asurin ring the Impac act t of 03 Demand Shaping Actio ions Record and measure the impact of demand shaping events such as promotions, price changes, product Downstream stream Data a launches and forward-buy 02 Integ egrat ratio ion arrangements. Include downstream supply chain data, such as POS data, in the demand forecasting model.

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Findings 3

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Characterization of Demand I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 15

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Characterization of Demand Shipments Time Series by Quarters 1 Day Mois oist 90-Pack 1 Day Mois oist for Asti tigm gmati tism 90-Pac ack 1 Day Mois oist 30-Pack 1 Day Mois oist for Asti tigm gmati tism 30 30-Pac ack 16

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Characterization of Demand Pareto Curves 1 Day Mois oist 90-Pack 1 Day Mois oist for Asti tigm gmati tism 90-Pac ack 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% % of Total SKUs % of Total SKUs 116 SKUs 1528 SKUs 1 Day Mois oist 30-Pack 1 Day Mois oist for Asti tigm gmati tism 30 30-Pac ack 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% % of Total SKUs % of Total SKUs 17

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Characterization of Demand Mean Shipments vs Coefficient of Variation for each Pareto segment 18

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Forecast Accuracy Analysis I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 19 Aggregation Levels Forecasts Compared Accuracy Metrics 1 2 3 Brand J&J Vision on Care - Statist stic ical 3-Month Average Bias — — — — Pack Size J&J Vision Care - Lag 03 4-Month Average MAPE — — — — Pareto to J&J Vision Care - Lag 02 5-Month Average MAPV — — — — SKU J&J Vision on Care - Lag 01 01 6-Month Average PVE — — — — Month Naïve Simple Exp. Smoothing RMSE — — — — Quarters 2-Mont nth h Average ge Double Exp. Smoothing — — —

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Forecast Accuracy Analysis Naïve vs Lag 01 Forecast Comparison Results 20

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Forecast Accuracy Analysis 2-Month Average vs Lag 01 Forecast Comparison Results 21

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Current Forecasting Process I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 22 1-Month Cycle 2 weeks 3 Shipments Data Executive S&OP Input 1 2 2.2 1.1 JDA 1.2 Regional 2.1 SKU Process Production Software S&OP Breakdown Planning Statistical Consensus SKU Level Output Forecast Forecast Forecast Lag 03 Lag 02 Lag 01

MIT Supply Chain Management Program Evan Humphrey | Federico Laiño Suggestions for Implementing Demand Sensing I N T R O D U C T I O N M E T H O D O L O G Y F I N D I N G S C O N C L U S I O N S 23 Latenc tency Reduct ction 01 Reduce cycle time between forecasts to take advantage of latest demand information updates. Meas asurin ring the Impac act t of 03 Demand Shaping Actio ions Record and measure the impact of demand shaping events such as promotions, price changes, product Downstream stream Data a launches and forward-buy 02 Integ egrat ratio ion arrangements. Include downstream supply chain data, such as POS data, in the demand forecasting model.