On the impact of inventory accuracy improvements on sales Christoph - PowerPoint PPT Presentation

On the impact of inventory accuracy improvements on sales Christoph Glock, Yacine Rekik, Aris A. Syntetos ECR Brussels – October 25, 2018 - 1 -

About us ■ Christoph Glock l Professor Chair: Technical University of Darmstadt, Germany l Specialises in Inventory Optimisation and Warehousing Management ■ Yacine Rekik l Professor Chair: EM-Lyon Business School, France l Specialises in Inventory Optimisation and Tracking (e.g. RFID) Technologies ■ Aris A. Syntetos l Panalpina Chaired Professor: Cardiff Business School, Cardiff University, UK l Specialises in Statistical Forecasting, Demand Classification & Inventory Optimisation. - 2 -

Background and objectives ■ Inventory inaccuracies: major issue in retailing and apparel industry. ■ Physical stock is (typically) less than what we think it is. l Most reasonable assumption in retailing. Generally, stores are negative in terms of stock. l Thus, reconciling inventories may only lead to an increase in sales. l ( We will see later that positive stock discrepancies are also possible, still leading though to reduced sales! ) ■ The problem has been established; we are not here to argue for its existence. - 3 -

Background and objectives ■ But rather: l Assess the implications of the problem, or rather the implications of fixing the problem (phase 1); l Assess alternative ways of fixing the problem itself (phase 2). ■ Phase 1: What is the impact on (increased) sales if inventory accuracy is increased by x%? l How does inventory accuracy develop over time after a stock take? l Is there an optimal number of stock takes? Do too many stock takes negatively influence inventory accuracy? l What exactly constitutes this problem of inventory discrepancies? ■ Phase 2 (upon clearly establishing the implications): what are the strategies to be employed (algorithmic driven, new identification technologies, counting, etc.) to fight the route causes of the problem? - 4 -

Error free ( r,Q ) inventory policy Order Quantity Q Q Reorder Point: r Safety Stock t L 2 L 1 Lead Time - 5 -

Impact of errors on the ( r,Q ) inventory policy This is the visible stock behavior: POS (Real Demand) + Replenishment Expected error free Order Quantity Q stock level Q Actual stock level subject to errors Reorder Point: r Safety Stock t L 2 L 1 Lead Time This is the invisible stock behavior: POS (Real Demand) + Replenishment + Skrink (Ghost Demand) - 6 -

Impact of errors on the ( r,Q ) inventory policy Wrong ordering Order Quantity Q Lost Sales are timing and more quantity frequent Q Errors act as “ghost” demand decreasing the stock level without generating a revenue: Inventory is controlled based on some visible wrong information, whereas sales are satisfied based on some Reorder Point: r correct but invisible information t Safety Stock L 2 L 1 Lead Time And more importantly, they are not detectable - 7 -

Experiment I: original / ideal experiment Stock Count Stock Count Stock Count Sales 12 weeks after the Sales (e.g.) 12 weeks “Test” inventory records have been before count “trued up” No Stock Count Stock Count Stock Count “Control” Sales 12 weeks after Sales (e.g.) over 12 weeks Stock Counts in the beginning and at the end of the experiment in both Test and Control stores - 8 -

In addition, Experiment II: retailers with frequent periodic stock audits several cases of Experiment I can be deduced from Experiment II Stock Count Stock Count Stock Count “Test” Stock Count Stock Count “Control” A comparison of sales on a same period of time, with different stock audit tasks in Test and Control stores - 9 -

Experiment III: weekly stock audit Week Week … 1 12 Stock Audit Physical Discrepancy: Computer Discrepancy: what was found vs. what what was found vs. what is Computer and Counted Physical stock during the should be in the stock shown in the computer audit are contrasted with what they should be given last week’s audit and before the audit stock movements (last counted physical stock + stock input – stock output) - 10 -

Empirical analysis ■ Currently working with 8 Retailers across Europe: l NDAs have been signed and we are in various phases with regards to data transfer and analysis; l 4 Grocery retailers (supermarkets), 2 Apparel retailers and 2 other; l Customised reports to be produced for all participating retailers. ■ Initial results: l 4 Grocery retailers (a, b, c and d) l +600,000 SKUs l 80 stores examined (40 test VS. 40 control) l We are still working on the best possible way to present descriptive statistics without unintentionally disclosing individual retailer information (and identity) l We present the (initial) results for the 4 Grocery retailers. - 11 -

Results PART I: GENERAL INSIGHTS (apply to all retailers) - 12 -

Result 1 Independently of the experimental setting, inventory record inaccuracies are an important issue for all participating companies: across all grocery retailers, between 46% and 73% of the audited SKUs are subject to inaccuracies even if a stock take is performed very frequently (as in Experiment II, or even each week as in Experiment III) - 13 -

Result 1 Independently of the experimental setting, inventory record inaccuracies are an important issue for all participating companies: across all grocery retailers, between 46% and 73% of the audited SKUs are subject to inaccuracies even if a stock take is performed very frequently (as in Experiment II or even each week, as in Experiment III) Stock Inaccuracy Level per Retailer 80% 73,33% 70% 63,25% 62,15% 60% 46,44% 50% 40% 30% 20% 10% 0% Retailer a Retailer b Retailer c Retailer d - 14 -

Result 1: inventory record inaccuracies constitute an important issue It is not only a matter of shrinkage: positive discrepancy is not negligible and generally is caused by Information System (IS) manipulations and errors 60% 53,57% 50% 44,34% 39,16% 37,80% 37,85% 40% 36,75% 28,99% 30% 26,67% 25,05% 24,35% 24,09% 21,39% 20% 10% 0% Retailer a Retailer b Retailer c Retailer d Negative Accurate Positive - 15 -

Result 2 Inventory record accuracy leads to an increase in sales turnover between 2.36% and 4.68% at the retailers. Sales Increase with better stock accuracy 5,0% 4,68% 4,5% 4,0% 3,5% 2,94% 2,70% 3,0% 2,36% 2,5% 2,0% 1,5% 1,0% 0,5% 0,0% Retailer a Retailer b Retailer c Retailer d - 16 -

Retailer a – sales increase of 2.94% in the Test stores comes from: 10% 6% 8,61% 4,86% 9% 5% 4,41% 8% 7,07% 3,79% 7% 4% 6% 5% 3% 4% 2,75% 2% 3% 2% 1% 1% 0% 0% Fast_Mover Middle_Mover Slow_Mover High_Discrepancy Middle_Discrepancy Slow_Discrepancy 7% 6,01% 6% 4,54% 5% 4% 3% 2% 0,67% 1% 0% Accurate Negative Positive - 17 -

Retailer b – sales increase of 2.36% in the Test stores comes from: 3% 2,26% 3,20% 4% 2,92% 3% 2% 3% 2% 2% 2% 1,16% 1% 0,62% 1% 1% 0,23% 1% 0% 0% Fast_Mover Middle_Mover Slow_Mover High_Discrepancy Middle_Discrepancy Slow_Discrepancy 5% 3,89% 4% 4% 3% 2,21% 3% 2% 2% 1% 1% 0% Accurate Negative Positive -1% -0,53% -1% - 18 -

Retailer c – sales increase of 2.70% in the Test stores comes from: 3,0% 2,58% 4,0% 3,62% 2,32% 2,5% 3,5% 2,12% 3,0% 2,0% 2,5% 1,5% 1,66% 2,0% 1,5% 1,0% 0,98% 1,0% 0,5% 0,5% 0,0% 0,0% Fast_Mover Middle_Mover Slow_Mover High_Discrepancy Middle_Discrepancy Slow_Discrepancy 2,82% 3,0% 2,35% 2,5% 2,0% 1,5% 1,0% 0,66% 0,5% 0,0% - 19 - Accurate Negative Positive

Retailer c: inaccuracy level per category ■ SKU Categories subject to both positive and negative discrepancies Inaccuracy Level per Category 90% 80% 70% 60% 50% 40% 30% Inaccuracy Level Negative Discrepancy 20% 10% 0% Y D E R Y T Y R T D Y Y E S R E O G E E A R O C E R R E L R T R T E R L C N I N E O U I E U H E T T E U T A O C I B R H K I L N M H A N D T C K T W F B A O I T U P A O U E A E L U O O B A M D N O O B U S B / O S B O R R C R O S E L C P U & d G C & C P T T G A C Z N n I O N T H O N H S H a A V H A R E r M V S R O T B L O L N O I F / R S P F A H I R W P W E P H E N ■ Can we infer that the sales increase offered by better stock accuracy is proportional to the inaccuracy level? - 20 -