SlideShare a Scribd company logo
Supply Optimization At Instacart
Jagannath Putrevu
SF Bay Area Machine Learning Meetup
5/3/2017
v
Supply Forecasting - McDonald’s
Single Server Queue
Service Time: 5 min/order
Orders per hour: 12
24 orders per hour
60 orders per hour
.
.
2 servers per hour
5 servers per hour
.
.
Demand Supply
v
Supply Forecasting - Chipotle?
Multi Server Queue
Variable Service Speeds Per Server
Demand Supply
24 orders per hour
60 order per hour
.
.
?
?
.
.
v
Supply Forecasting - Instacart?
What the talk will cover:
● Logistics @ Instacart
● Fulfillment Models
● Problem Formulation
● Simulation Based Optimization Framework
What it will not cover:
● Technology stack
● Algorithm details
● Results & Comparisons
v
Delivery Logistics
Customers Place
Orders
Orders Get
Shopped For
Orders Get
Delivered
v
Fulfillment Models
Due
#2 Ordered
Due#1 Ordered
Due#3 Ordered
Handoff
Model
Shopping Times Delivery Times
#1
#2 #3
#1 #3 #2
In-Store Shopper 1
In-Store Shopper 2
Driver 1
Handoff
Pickup
Bags
v
Handoff Model - Staging Area
v
Fulfillment Models
Full Service
Model
Shopping Times Delivery Times
Due
#2 Ordered
Due#1 Ordered
Due#3 Ordered
Full Service Shopper 1 #3 #2#1#1 #2 #3
Store With Personal Shoppers
Full Service Shopper
Customer
Store With Personal Shoppers
Full Service Shopper
Customer
v
What is Efficiency?
Efficiency =
Deliveries
Total Time
Deliveries
Active Time
Active Time
Total Time
= X
Active Efficiency= X Utilization
How fast? How busy?
v
Time Components In Labor Hours
Begin Shift End Shift
Deliver to Customer
Wait for Batch
Drive to Store Shop Groceries
Pickup Bags from Store
v
Time Components In Labor Hours
Deliver to Customer
Wait for Batch
Drive to Store Shop Groceries
Pickup Bags from Store
End ShiftBegin Shift
In-store Shopper
v
Time Components In Labor Hours
Deliver to Customer
Wait for Batch
Drive to Store Shop Groceries
Pickup Bags from Store
Full Service
Begin Shift
FullServiceShopper
FullService
FullService
End Shift
v
Time Components In Labor Hours
Deliver to Customer
Wait for Batch
Drive to Store Shop Groceries
Pickup Bags from Store
Driver
DriverDriverBegin Shift
Driver
End Shift
v
Time Components And Factors Affecting Them
Idle Time: Supply/Demand equilibrium, Variance in Cancellations
How busy?
How fast?
Driving Time: Batching (Number of Orders Per Trip), Proximity To Store, Traffic
Picking Time: Speed of Shopper, Shopping List Sorting, Checkout Times
Bags Pickup Time: Staging Area Layout, Number of Bags being picked up
Delivery Time: Traffic, Order Time/Space Density, Routing Algorithm Efficiency
v
Supply/Demand Equilibrium
Idle Time
Lost Deliveries
v
Problem Statement
● For a given day, how do we staff :
○ In-store shoppers (ISS) at each store location,
○ Full Service shoppers (FSS) across the entire zone,
● so as to:
○ maximize active efficiency,
○ minimize idleness,
○ minimize lost deliveries
v
Final goal?
8 AM 9 AM 10 AM 11 AM 12 PM 1 PM 2 PM 3 PM 4 PM 5 PM 6 PM 7 PM 8 PM
Whole Foods
Soma
2 3 4 5 6 6 6 5 5 5 4 3 2
Whole Foods
Castro
? ? ? ? ? ? ? ? ? ? ? ? ?
Costco ? ? ? ? ? ? ? ? ? ? ? ? ?
Safeway ? ? ? ? ? ? ? ? ? ? ? ? ?
Full Service 20 30 40 50 60 60 60 50 50 50 40 30 20
v
Decision Factors
Handoff Model Full Service Model
Speed
Location
Overhead
Task Handling
v
Everything Has Variance
● Demand
● Space-Time Density
● Traffic
● Weather
● Shopper Abilities
● Cancellations
v
Cost Functions
Desired Actual
Idleness Cost
Lost Deliveries Cost
Lost Efficiency Cost
v
System
Outputs
Number of Shoppers Required
Percentage of Volume Done Through
Handoff
Estimated Efficiency & Utilization
Inputs
Demand
(variable)
Store Locations
(fixed)
Shopper Pool
(variable)
Fulfillment Times
(predicted, variable)
Cancellation Probability
(predicted)
Handoff vs Full-Service
Idleness vs Efficiency vs Lost
Deliveries
Store Timings Shift Length Requirements Business rules
Constraints
v
Simulate An Ideal Universe
Assume infinite Supply, solve the fulfillment problem, and generate optimal set of trips
v
Staff For An Ideal Universe
● When do all the trips begin and end in the optimal solution?
● Back-calculate how many shoppers we need and at what times we need them
v
Results From One Universe
v
Aggregate Results From All Simulations
v
Final ‘Optimal’ Set of Staffing Levels
v
Losses In One Universe
1
1
1
2
2
2
1
2
Idleness Cost
Lost Efficiency Cost
(and/or)
Lost Deliveries Cost
x_8
x_9
x_10
x_11
x_12
x_13
x_14
x_15
x_16
x_17
x_18
x_19
x_20
v
Minimize Losses Across All Universes
Min. ∑ (Idleness Cost + Lost Efficiency Cost + Lost Deliveries Cost)
v
Final Output
8 AM 9 AM 10 AM 11 AM 12 PM 1 PM 2 PM 3 PM 4 PM 5 PM 6 PM 7 PM 8 PM
Whole Foods
Soma
12 12 16 16 14 14 12 12 10 10 8 6 4
Whole Foods
Castro
5 5 5 5 4 4 4 3 3 3 3 2 0
Costco 0 0 10 10 10 8 8 8 6 6 6 0 0
Safeway 2 3 4 5 6 6 6 5 5 5 4 3 2
Full Service 20 60 100 120 150 180 180 180 150 150 80 50 30

More Related Content

What's hot

Prepaid card in india
Prepaid card in indiaPrepaid card in india
Prepaid card in india
MorCryp
 
IBM Payments Gateway
IBM Payments GatewayIBM Payments Gateway
IBM Payments Gateway
Killian Delaney
 
E wallet
E walletE wallet
E wallet
anuja joshi
 
PayPal - Strategy Analysis & Frameworks
PayPal - Strategy Analysis & FrameworksPayPal - Strategy Analysis & Frameworks
PayPal - Strategy Analysis & Frameworks
Masoom Modh
 
Analystics in banking and financial services
Analystics in banking and financial servicesAnalystics in banking and financial services
Analystics in banking and financial services
RoshithaSunil
 
Mobile Wallet functions
Mobile Wallet functionsMobile Wallet functions
Mobile Wallet functions
Mikhail Miroshnichenko
 
EJ Management by ESQ
EJ Management by ESQEJ Management by ESQ
EJ Management by ESQ
ESQ Business Services
 
Payment Gateway
Payment GatewayPayment Gateway
Payment Gateway
Nyros Technologies
 
gay indonesia dan video gay indonesia
gay indonesia dan video gay indonesia gay indonesia dan video gay indonesia
gay indonesia dan video gay indonesia
aseadi
 
How does Instacart Works
How does Instacart WorksHow does Instacart Works
How does Instacart Works
Infigic Digital Solutions
 
Optimization Algorithms for Meal Delivery Operations
Optimization Algorithms for Meal Delivery OperationsOptimization Algorithms for Meal Delivery Operations
Optimization Algorithms for Meal Delivery Operations
Alan Erera
 
eCommerce Payment Gateways: An Introduction
eCommerce Payment Gateways: An IntroductioneCommerce Payment Gateways: An Introduction
eCommerce Payment Gateways: An Introduction
AidanChard
 
4 way recon solution for ATM,POS,Recyclers,Mobile banking, Internet banking,etc
4 way recon solution for ATM,POS,Recyclers,Mobile banking, Internet banking,etc4 way recon solution for ATM,POS,Recyclers,Mobile banking, Internet banking,etc
4 way recon solution for ATM,POS,Recyclers,Mobile banking, Internet banking,etc
Pratap Parab
 
Artificial Intelligence in Banking
Artificial Intelligence in BankingArtificial Intelligence in Banking
Artificial Intelligence in Banking
Khawar Nehal khawar.nehal@atrc.net.pk
 
Atm
AtmAtm
Customer Churn Prevention Powerpoint Presentation Slides
Customer Churn Prevention Powerpoint Presentation SlidesCustomer Churn Prevention Powerpoint Presentation Slides
Customer Churn Prevention Powerpoint Presentation Slides
SlideTeam
 
AI powered decision making in banks
AI powered decision making in banksAI powered decision making in banks
AI powered decision making in banks
Pankaj Baid
 
Mobile Payments - How is it done?
Mobile Payments - How is it done?Mobile Payments - How is it done?
Mobile Payments - How is it done?
Parag Arjunwadkar
 
Instacart
InstacartInstacart
Instacart
Nino Panes
 

What's hot (20)

Prepaid card in india
Prepaid card in indiaPrepaid card in india
Prepaid card in india
 
IBM Payments Gateway
IBM Payments GatewayIBM Payments Gateway
IBM Payments Gateway
 
E wallet
E walletE wallet
E wallet
 
PayPal - Strategy Analysis & Frameworks
PayPal - Strategy Analysis & FrameworksPayPal - Strategy Analysis & Frameworks
PayPal - Strategy Analysis & Frameworks
 
Analystics in banking and financial services
Analystics in banking and financial servicesAnalystics in banking and financial services
Analystics in banking and financial services
 
Mobile Wallet functions
Mobile Wallet functionsMobile Wallet functions
Mobile Wallet functions
 
EJ Management by ESQ
EJ Management by ESQEJ Management by ESQ
EJ Management by ESQ
 
Payment Gateway
Payment GatewayPayment Gateway
Payment Gateway
 
gay indonesia dan video gay indonesia
gay indonesia dan video gay indonesia gay indonesia dan video gay indonesia
gay indonesia dan video gay indonesia
 
How does Instacart Works
How does Instacart WorksHow does Instacart Works
How does Instacart Works
 
Optimization Algorithms for Meal Delivery Operations
Optimization Algorithms for Meal Delivery OperationsOptimization Algorithms for Meal Delivery Operations
Optimization Algorithms for Meal Delivery Operations
 
eCommerce Payment Gateways: An Introduction
eCommerce Payment Gateways: An IntroductioneCommerce Payment Gateways: An Introduction
eCommerce Payment Gateways: An Introduction
 
4 way recon solution for ATM,POS,Recyclers,Mobile banking, Internet banking,etc
4 way recon solution for ATM,POS,Recyclers,Mobile banking, Internet banking,etc4 way recon solution for ATM,POS,Recyclers,Mobile banking, Internet banking,etc
4 way recon solution for ATM,POS,Recyclers,Mobile banking, Internet banking,etc
 
Artificial Intelligence in Banking
Artificial Intelligence in BankingArtificial Intelligence in Banking
Artificial Intelligence in Banking
 
Atm
AtmAtm
Atm
 
Customer Churn Prevention Powerpoint Presentation Slides
Customer Churn Prevention Powerpoint Presentation SlidesCustomer Churn Prevention Powerpoint Presentation Slides
Customer Churn Prevention Powerpoint Presentation Slides
 
AI powered decision making in banks
AI powered decision making in banksAI powered decision making in banks
AI powered decision making in banks
 
Mobile Payments - How is it done?
Mobile Payments - How is it done?Mobile Payments - How is it done?
Mobile Payments - How is it done?
 
PayPal
PayPal PayPal
PayPal
 
Instacart
InstacartInstacart
Instacart
 

Similar to Supply optimization instacart

Real Time Fulfilment Planning
Real Time Fulfilment PlanningReal Time Fulfilment Planning
Real Time Fulfilment PlanningJagadeesh Huliyar
 
Future group iift the strategists
Future group iift the strategistsFuture group iift the strategists
Future group iift the strategists
Gunjan Solanki
 
Mli 2017 business navigating mcom
Mli 2017 business navigating mcomMli 2017 business navigating mcom
Mli 2017 business navigating mcom
Hanoi MagentoMeetup
 
OM Week 1.pptx
OM Week 1.pptxOM Week 1.pptx
OM Week 1.pptx
NamanAgarwal323573
 
Sales Force & Merchandiser Enablement Through Mobility
Sales Force & Merchandiser Enablement Through MobilitySales Force & Merchandiser Enablement Through Mobility
Sales Force & Merchandiser Enablement Through MobilityThe Information Company
 
Sales Force & Merchandiser Enablement Through Mobility
Sales Force & Merchandiser Enablement Through MobilitySales Force & Merchandiser Enablement Through Mobility
Sales Force & Merchandiser Enablement Through Mobility
The Information Company
 
How Euroflorist is preparing for Artificial Intelligence
How Euroflorist is preparing for Artificial IntelligenceHow Euroflorist is preparing for Artificial Intelligence
How Euroflorist is preparing for Artificial Intelligence
Guido X Jansen
 
Preparing for AI - Measurefest
Preparing for AI - MeasurefestPreparing for AI - Measurefest
Preparing for AI - Measurefest
Guido X Jansen
 
Uop ops-571-final-exam-guide- april-2018--new
Uop ops-571-final-exam-guide- april-2018--newUop ops-571-final-exam-guide- april-2018--new
Uop ops-571-final-exam-guide- april-2018--new
indi7257
 
How Gousto is moving to just-in-time personalization with Snowplow
How Gousto is moving to just-in-time personalization with SnowplowHow Gousto is moving to just-in-time personalization with Snowplow
How Gousto is moving to just-in-time personalization with Snowplow
Giuseppe Gaviani
 
Pepsi sales force enablement & merchandising through mobility
Pepsi  sales force enablement & merchandising through mobilityPepsi  sales force enablement & merchandising through mobility
Pepsi sales force enablement & merchandising through mobilityThe Information Company
 
Food delivery - Supply Chain Logistics Model & Frame work
Food delivery - Supply Chain Logistics Model & Frame workFood delivery - Supply Chain Logistics Model & Frame work
Food delivery - Supply Chain Logistics Model & Frame work
Alvis Lazarus
 
Manufacturing's Holy Grail: A Practical Science for Executives and Managers
Manufacturing's Holy Grail: A Practical Science for Executives and ManagersManufacturing's Holy Grail: A Practical Science for Executives and Managers
Manufacturing's Holy Grail: A Practical Science for Executives and Managers
UBMCanon
 
Supply Chain Analytics with Simulation
Supply Chain Analytics with SimulationSupply Chain Analytics with Simulation
Supply Chain Analytics with Simulation
Steve Haekler
 
Supply Chain Analytics with Simulation
Supply Chain Analytics with SimulationSupply Chain Analytics with Simulation
Supply Chain Analytics with Simulation
ProModel Corporation
 
Queuing Theory.pdf
Queuing Theory.pdfQueuing Theory.pdf
Queuing Theory.pdf
Erandika Gamage
 
DT Smart checkouts.pptx
DT Smart checkouts.pptxDT Smart checkouts.pptx
DT Smart checkouts.pptx
20PA013BHOOMIKAP
 
Omni channel fulfilment and supply chain management analytic
Omni channel fulfilment and supply chain management analyticOmni channel fulfilment and supply chain management analytic
Omni channel fulfilment and supply chain management analytic
Amit Kumar Garg
 
Data Wrangling Questionnaire
Data Wrangling QuestionnaireData Wrangling Questionnaire
Data Wrangling Questionnaire
Mimi Brown
 

Similar to Supply optimization instacart (20)

Real Time Fulfilment Planning
Real Time Fulfilment PlanningReal Time Fulfilment Planning
Real Time Fulfilment Planning
 
Future group iift the strategists
Future group iift the strategistsFuture group iift the strategists
Future group iift the strategists
 
Mli 2017 business navigating mcom
Mli 2017 business navigating mcomMli 2017 business navigating mcom
Mli 2017 business navigating mcom
 
OM Week 1.pptx
OM Week 1.pptxOM Week 1.pptx
OM Week 1.pptx
 
Sales Force & Merchandiser Enablement Through Mobility
Sales Force & Merchandiser Enablement Through MobilitySales Force & Merchandiser Enablement Through Mobility
Sales Force & Merchandiser Enablement Through Mobility
 
Sales Force & Merchandiser Enablement Through Mobility
Sales Force & Merchandiser Enablement Through MobilitySales Force & Merchandiser Enablement Through Mobility
Sales Force & Merchandiser Enablement Through Mobility
 
Creative Concept
Creative ConceptCreative Concept
Creative Concept
 
How Euroflorist is preparing for Artificial Intelligence
How Euroflorist is preparing for Artificial IntelligenceHow Euroflorist is preparing for Artificial Intelligence
How Euroflorist is preparing for Artificial Intelligence
 
Preparing for AI - Measurefest
Preparing for AI - MeasurefestPreparing for AI - Measurefest
Preparing for AI - Measurefest
 
Uop ops-571-final-exam-guide- april-2018--new
Uop ops-571-final-exam-guide- april-2018--newUop ops-571-final-exam-guide- april-2018--new
Uop ops-571-final-exam-guide- april-2018--new
 
How Gousto is moving to just-in-time personalization with Snowplow
How Gousto is moving to just-in-time personalization with SnowplowHow Gousto is moving to just-in-time personalization with Snowplow
How Gousto is moving to just-in-time personalization with Snowplow
 
Pepsi sales force enablement & merchandising through mobility
Pepsi  sales force enablement & merchandising through mobilityPepsi  sales force enablement & merchandising through mobility
Pepsi sales force enablement & merchandising through mobility
 
Food delivery - Supply Chain Logistics Model & Frame work
Food delivery - Supply Chain Logistics Model & Frame workFood delivery - Supply Chain Logistics Model & Frame work
Food delivery - Supply Chain Logistics Model & Frame work
 
Manufacturing's Holy Grail: A Practical Science for Executives and Managers
Manufacturing's Holy Grail: A Practical Science for Executives and ManagersManufacturing's Holy Grail: A Practical Science for Executives and Managers
Manufacturing's Holy Grail: A Practical Science for Executives and Managers
 
Supply Chain Analytics with Simulation
Supply Chain Analytics with SimulationSupply Chain Analytics with Simulation
Supply Chain Analytics with Simulation
 
Supply Chain Analytics with Simulation
Supply Chain Analytics with SimulationSupply Chain Analytics with Simulation
Supply Chain Analytics with Simulation
 
Queuing Theory.pdf
Queuing Theory.pdfQueuing Theory.pdf
Queuing Theory.pdf
 
DT Smart checkouts.pptx
DT Smart checkouts.pptxDT Smart checkouts.pptx
DT Smart checkouts.pptx
 
Omni channel fulfilment and supply chain management analytic
Omni channel fulfilment and supply chain management analyticOmni channel fulfilment and supply chain management analytic
Omni channel fulfilment and supply chain management analytic
 
Data Wrangling Questionnaire
Data Wrangling QuestionnaireData Wrangling Questionnaire
Data Wrangling Questionnaire
 

Recently uploaded

一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单
一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单
一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单
ewymefz
 
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
NABLAS株式会社
 
社内勉強会資料_LLM Agents                              .
社内勉強会資料_LLM Agents                              .社内勉強会資料_LLM Agents                              .
社内勉強会資料_LLM Agents                              .
NABLAS株式会社
 
SOCRadar Germany 2024 Threat Landscape Report
SOCRadar Germany 2024 Threat Landscape ReportSOCRadar Germany 2024 Threat Landscape Report
SOCRadar Germany 2024 Threat Landscape Report
SOCRadar
 
Predicting Product Ad Campaign Performance: A Data Analysis Project Presentation
Predicting Product Ad Campaign Performance: A Data Analysis Project PresentationPredicting Product Ad Campaign Performance: A Data Analysis Project Presentation
Predicting Product Ad Campaign Performance: A Data Analysis Project Presentation
Boston Institute of Analytics
 
Chatty Kathy - UNC Bootcamp Final Project Presentation - Final Version - 5.23...
Chatty Kathy - UNC Bootcamp Final Project Presentation - Final Version - 5.23...Chatty Kathy - UNC Bootcamp Final Project Presentation - Final Version - 5.23...
Chatty Kathy - UNC Bootcamp Final Project Presentation - Final Version - 5.23...
John Andrews
 
一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单
ewymefz
 
一比一原版(Deakin毕业证书)迪肯大学毕业证如何办理
一比一原版(Deakin毕业证书)迪肯大学毕业证如何办理一比一原版(Deakin毕业证书)迪肯大学毕业证如何办理
一比一原版(Deakin毕业证书)迪肯大学毕业证如何办理
oz8q3jxlp
 
一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单
一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单
一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单
vcaxypu
 
哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样
哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样
哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样
axoqas
 
Levelwise PageRank with Loop-Based Dead End Handling Strategy : SHORT REPORT ...
Levelwise PageRank with Loop-Based Dead End Handling Strategy : SHORT REPORT ...Levelwise PageRank with Loop-Based Dead End Handling Strategy : SHORT REPORT ...
Levelwise PageRank with Loop-Based Dead End Handling Strategy : SHORT REPORT ...
Subhajit Sahu
 
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
nscud
 
一比一原版(UofS毕业证书)萨省大学毕业证如何办理
一比一原版(UofS毕业证书)萨省大学毕业证如何办理一比一原版(UofS毕业证书)萨省大学毕业证如何办理
一比一原版(UofS毕业证书)萨省大学毕业证如何办理
v3tuleee
 
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...
pchutichetpong
 
Malana- Gimlet Market Analysis (Portfolio 2)
Malana- Gimlet Market Analysis (Portfolio 2)Malana- Gimlet Market Analysis (Portfolio 2)
Malana- Gimlet Market Analysis (Portfolio 2)
TravisMalana
 
一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单
一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单
一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单
ewymefz
 
一比一原版(TWU毕业证)西三一大学毕业证成绩单
一比一原版(TWU毕业证)西三一大学毕业证成绩单一比一原版(TWU毕业证)西三一大学毕业证成绩单
一比一原版(TWU毕业证)西三一大学毕业证成绩单
ocavb
 
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
u86oixdj
 
Algorithmic optimizations for Dynamic Levelwise PageRank (from STICD) : SHORT...
Algorithmic optimizations for Dynamic Levelwise PageRank (from STICD) : SHORT...Algorithmic optimizations for Dynamic Levelwise PageRank (from STICD) : SHORT...
Algorithmic optimizations for Dynamic Levelwise PageRank (from STICD) : SHORT...
Subhajit Sahu
 
Best best suvichar in gujarati english meaning of this sentence as Silk road ...
Best best suvichar in gujarati english meaning of this sentence as Silk road ...Best best suvichar in gujarati english meaning of this sentence as Silk road ...
Best best suvichar in gujarati english meaning of this sentence as Silk road ...
AbhimanyuSinha9
 

Recently uploaded (20)

一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单
一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单
一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单
 
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
 
社内勉強会資料_LLM Agents                              .
社内勉強会資料_LLM Agents                              .社内勉強会資料_LLM Agents                              .
社内勉強会資料_LLM Agents                              .
 
SOCRadar Germany 2024 Threat Landscape Report
SOCRadar Germany 2024 Threat Landscape ReportSOCRadar Germany 2024 Threat Landscape Report
SOCRadar Germany 2024 Threat Landscape Report
 
Predicting Product Ad Campaign Performance: A Data Analysis Project Presentation
Predicting Product Ad Campaign Performance: A Data Analysis Project PresentationPredicting Product Ad Campaign Performance: A Data Analysis Project Presentation
Predicting Product Ad Campaign Performance: A Data Analysis Project Presentation
 
Chatty Kathy - UNC Bootcamp Final Project Presentation - Final Version - 5.23...
Chatty Kathy - UNC Bootcamp Final Project Presentation - Final Version - 5.23...Chatty Kathy - UNC Bootcamp Final Project Presentation - Final Version - 5.23...
Chatty Kathy - UNC Bootcamp Final Project Presentation - Final Version - 5.23...
 
一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单
 
一比一原版(Deakin毕业证书)迪肯大学毕业证如何办理
一比一原版(Deakin毕业证书)迪肯大学毕业证如何办理一比一原版(Deakin毕业证书)迪肯大学毕业证如何办理
一比一原版(Deakin毕业证书)迪肯大学毕业证如何办理
 
一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单
一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单
一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单
 
哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样
哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样
哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样
 
Levelwise PageRank with Loop-Based Dead End Handling Strategy : SHORT REPORT ...
Levelwise PageRank with Loop-Based Dead End Handling Strategy : SHORT REPORT ...Levelwise PageRank with Loop-Based Dead End Handling Strategy : SHORT REPORT ...
Levelwise PageRank with Loop-Based Dead End Handling Strategy : SHORT REPORT ...
 
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
 
一比一原版(UofS毕业证书)萨省大学毕业证如何办理
一比一原版(UofS毕业证书)萨省大学毕业证如何办理一比一原版(UofS毕业证书)萨省大学毕业证如何办理
一比一原版(UofS毕业证书)萨省大学毕业证如何办理
 
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...
 
Malana- Gimlet Market Analysis (Portfolio 2)
Malana- Gimlet Market Analysis (Portfolio 2)Malana- Gimlet Market Analysis (Portfolio 2)
Malana- Gimlet Market Analysis (Portfolio 2)
 
一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单
一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单
一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单
 
一比一原版(TWU毕业证)西三一大学毕业证成绩单
一比一原版(TWU毕业证)西三一大学毕业证成绩单一比一原版(TWU毕业证)西三一大学毕业证成绩单
一比一原版(TWU毕业证)西三一大学毕业证成绩单
 
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
 
Algorithmic optimizations for Dynamic Levelwise PageRank (from STICD) : SHORT...
Algorithmic optimizations for Dynamic Levelwise PageRank (from STICD) : SHORT...Algorithmic optimizations for Dynamic Levelwise PageRank (from STICD) : SHORT...
Algorithmic optimizations for Dynamic Levelwise PageRank (from STICD) : SHORT...
 
Best best suvichar in gujarati english meaning of this sentence as Silk road ...
Best best suvichar in gujarati english meaning of this sentence as Silk road ...Best best suvichar in gujarati english meaning of this sentence as Silk road ...
Best best suvichar in gujarati english meaning of this sentence as Silk road ...
 

Supply optimization instacart

  • 1. Supply Optimization At Instacart Jagannath Putrevu SF Bay Area Machine Learning Meetup 5/3/2017
  • 2. v Supply Forecasting - McDonald’s Single Server Queue Service Time: 5 min/order Orders per hour: 12 24 orders per hour 60 orders per hour . . 2 servers per hour 5 servers per hour . . Demand Supply
  • 3. v Supply Forecasting - Chipotle? Multi Server Queue Variable Service Speeds Per Server Demand Supply 24 orders per hour 60 order per hour . . ? ? . .
  • 4. v Supply Forecasting - Instacart? What the talk will cover: ● Logistics @ Instacart ● Fulfillment Models ● Problem Formulation ● Simulation Based Optimization Framework What it will not cover: ● Technology stack ● Algorithm details ● Results & Comparisons
  • 5. v Delivery Logistics Customers Place Orders Orders Get Shopped For Orders Get Delivered
  • 6. v Fulfillment Models Due #2 Ordered Due#1 Ordered Due#3 Ordered Handoff Model Shopping Times Delivery Times #1 #2 #3 #1 #3 #2 In-Store Shopper 1 In-Store Shopper 2 Driver 1 Handoff Pickup Bags
  • 7. v Handoff Model - Staging Area
  • 8. v Fulfillment Models Full Service Model Shopping Times Delivery Times Due #2 Ordered Due#1 Ordered Due#3 Ordered Full Service Shopper 1 #3 #2#1#1 #2 #3
  • 9. Store With Personal Shoppers Full Service Shopper Customer
  • 10. Store With Personal Shoppers Full Service Shopper Customer
  • 11. v What is Efficiency? Efficiency = Deliveries Total Time Deliveries Active Time Active Time Total Time = X Active Efficiency= X Utilization How fast? How busy?
  • 12. v Time Components In Labor Hours Begin Shift End Shift Deliver to Customer Wait for Batch Drive to Store Shop Groceries Pickup Bags from Store
  • 13. v Time Components In Labor Hours Deliver to Customer Wait for Batch Drive to Store Shop Groceries Pickup Bags from Store End ShiftBegin Shift In-store Shopper
  • 14. v Time Components In Labor Hours Deliver to Customer Wait for Batch Drive to Store Shop Groceries Pickup Bags from Store Full Service Begin Shift FullServiceShopper FullService FullService End Shift
  • 15. v Time Components In Labor Hours Deliver to Customer Wait for Batch Drive to Store Shop Groceries Pickup Bags from Store Driver DriverDriverBegin Shift Driver End Shift
  • 16. v Time Components And Factors Affecting Them Idle Time: Supply/Demand equilibrium, Variance in Cancellations How busy? How fast? Driving Time: Batching (Number of Orders Per Trip), Proximity To Store, Traffic Picking Time: Speed of Shopper, Shopping List Sorting, Checkout Times Bags Pickup Time: Staging Area Layout, Number of Bags being picked up Delivery Time: Traffic, Order Time/Space Density, Routing Algorithm Efficiency
  • 18. v Problem Statement ● For a given day, how do we staff : ○ In-store shoppers (ISS) at each store location, ○ Full Service shoppers (FSS) across the entire zone, ● so as to: ○ maximize active efficiency, ○ minimize idleness, ○ minimize lost deliveries
  • 19. v Final goal? 8 AM 9 AM 10 AM 11 AM 12 PM 1 PM 2 PM 3 PM 4 PM 5 PM 6 PM 7 PM 8 PM Whole Foods Soma 2 3 4 5 6 6 6 5 5 5 4 3 2 Whole Foods Castro ? ? ? ? ? ? ? ? ? ? ? ? ? Costco ? ? ? ? ? ? ? ? ? ? ? ? ? Safeway ? ? ? ? ? ? ? ? ? ? ? ? ? Full Service 20 30 40 50 60 60 60 50 50 50 40 30 20
  • 20. v Decision Factors Handoff Model Full Service Model Speed Location Overhead Task Handling
  • 21. v Everything Has Variance ● Demand ● Space-Time Density ● Traffic ● Weather ● Shopper Abilities ● Cancellations
  • 22. v Cost Functions Desired Actual Idleness Cost Lost Deliveries Cost Lost Efficiency Cost
  • 23. v System Outputs Number of Shoppers Required Percentage of Volume Done Through Handoff Estimated Efficiency & Utilization Inputs Demand (variable) Store Locations (fixed) Shopper Pool (variable) Fulfillment Times (predicted, variable) Cancellation Probability (predicted) Handoff vs Full-Service Idleness vs Efficiency vs Lost Deliveries Store Timings Shift Length Requirements Business rules Constraints
  • 24. v Simulate An Ideal Universe Assume infinite Supply, solve the fulfillment problem, and generate optimal set of trips
  • 25. v Staff For An Ideal Universe ● When do all the trips begin and end in the optimal solution? ● Back-calculate how many shoppers we need and at what times we need them
  • 26. v Results From One Universe
  • 27. v Aggregate Results From All Simulations
  • 28. v Final ‘Optimal’ Set of Staffing Levels
  • 29. v Losses In One Universe 1 1 1 2 2 2 1 2 Idleness Cost Lost Efficiency Cost (and/or) Lost Deliveries Cost x_8 x_9 x_10 x_11 x_12 x_13 x_14 x_15 x_16 x_17 x_18 x_19 x_20
  • 30. v Minimize Losses Across All Universes Min. ∑ (Idleness Cost + Lost Efficiency Cost + Lost Deliveries Cost)
  • 31. v Final Output 8 AM 9 AM 10 AM 11 AM 12 PM 1 PM 2 PM 3 PM 4 PM 5 PM 6 PM 7 PM 8 PM Whole Foods Soma 12 12 16 16 14 14 12 12 10 10 8 6 4 Whole Foods Castro 5 5 5 5 4 4 4 3 3 3 3 2 0 Costco 0 0 10 10 10 8 8 8 6 6 6 0 0 Safeway 2 3 4 5 6 6 6 5 5 5 4 3 2 Full Service 20 60 100 120 150 180 180 180 150 150 80 50 30