Sply Chain ModManf&Ware
Sply Chain ModManf&Ware ISYE 3104
Popular in Course
Popular in Industrial Engineering
This 0 page Class Notes was uploaded by Maryse Thiel on Monday November 2, 2015. The Class Notes belongs to ISYE 3104 at Georgia Institute of Technology - Main Campus taught by Staff in Fall. Since its upload, it has received 7 views. For similar materials see /class/234210/isye-3104-georgia-institute-of-technology-main-campus in Industrial Engineering at Georgia Institute of Technology - Main Campus.
Reviews for Sply Chain ModManf&Ware
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 11/02/15
Warehousing Systems Most ofthese slides are courtesy ofProfessor Paul Grif n Why Have a Warehouse Warehouses require many expensive resources capital Hlabor Pa facilities iE39 management systems Pa etc November 21 2007 Shipping directly from the suppliers retailer Direct from farms Potential problems a outof stock items late deliveries multiple items from the same order arrive at different times tracking tracing and returns are more complicated November 21 2007 Distribution from a centralized warehouse Inventory status FedEx dedicated edistributiori facility in Memphis November 21 2007 Distribution from multiple warehouses Peapod Amazoncom lE39 300 million distributioncenter initiative giant facilities in Nevada Kentucky and Kansas Barnesandnoblecom Pros and cons a High investment costs a Shorter delivery times Product Consolidation There is a cost associated with each product movement Distributor can ix consolidate shipments i for downstream 1 customers can reduce transportation costs can facilitate downstream receiving November 21 2007 Product Positioning I Position product close to the A customer Speed 7 Economies of Scale To realize economies of scale from price breaks purchasing To realize economies of scale in storage equipment space November 21 2007 Product Buffer Can provide and inventory buffer for evariability in demand seasonality of demand Value Adding Some warehouses are used to add value to the product Ea packaging assembly Postponement November 21 2007 Warehousing Evolution The role of warehousing has evolved over the years 19501970 primary role of the warehouse was the storage function 19701990 the rise of distribution centers DC which included order assembly as a key component 1990present the rise of the logistics center which included value added services on top of the DC functions Storage though still important is not the key driver that it once was Warehousing Functions r 913 W Order Pick Items and storage equipment in a warehouse Case Pallet Piece Gravity ow rack Bin shelving Carousel 13 Current Key Warehousing Issues Increased use of cross docking Increased use of fastpick Increased use of value added services and customization Need to deal with reverse logistics eg returns Increase in complexity eg SKU proliferation Increased accuracy requirements November 21 2007 November 21 2007 Warehousing Objective Minimize the set of costs elabor iE39 space E capital e1 IT etc Subject to constraints from customer e1 fill rate e1 response time e1 accuracy November 21 2007 What is activity profiling A systematic analysis of the activity of order and item activity during warehousing operations It helps us to identify EPolicyimprovement opportunities e b fd s 3 Z 39 e3 FE Equnpment selectionuse changes 8 A9 a 3 FE Layout redesign opportunities 39 3 9 FE Key SKUs Y 1 1quot The data IS key Pick sheet SKU Quantity Location 026596 4 278538 1 093478 22 Number of lines or line items or pick lines 3 Number of picks 3 Number of grabs 27 Example statistics Average number of E39 shipments per day E39 SKUs in the warehouse E39 lines per order E39 units pieces cases per line E39 orderpickers devoted to pallet movement casepicking and to brokencase picking Seasonalities I etc November 21 2007 Basics of profiling Important to realize that there is a tradeoff between E39 Simple aggregate statistical description E39 Complex description Simple statistics can give a quick albeit rough description of the warehouse However looking at averages can hide complexity including correlation confounding etc Example For our warehouse the average lines per order is 10 If we look at the actual distribution of lines per order we see the following El 50 of the time the lines per order is 1 El 50 of the time the lines per order is 19 Timebased analysis Several things change for a warehouse of time iE39 weekly seasonality of SKUs iE39 monthly seasonality of total demand iE39 daily seasonality of orders Issue a warehouse needs to be designed to handle the peaks not just the average iE39 opportunity to reduce some of these requirements by properly planning ahead November 21 2007 November 21 2007 Monthly Loads 20 quotquotquot Avg 10 f f V 0 1 1 1 1 1 1 1 2 3 4 5 6 7 8 9 10 11 12 Variabilitybased profiling Often times when we report averages we smooth over the true variability of the process Example Time Picks 800900 110 9001000 40 10001100 33 11001200 35 1200100 29 100200 180 200300 140 November 21 2007 Example continued For this example if we staffed to the average then we would need to be able to handle 81 picks per hour However the standard deviation for this process is very high gt61 and if we designed to the average then for 3 of the hour periods we would be SEVERELY underdesigned Categorization of items There are several different ways in which SKUs can be ordered an based on dollar annual volume e1 based on case movements e1 based on number of times picked e1 based on weight moved E39 etc Each of these can give a very different picture of warehouse operations and ABC classification of items November 21 2007 ABC classification Number of SKUs 510 A 50 45 50 B 450 40 50 7 0 4 5 IUQIIIISQAUI IBHIIIIV Top 10 items by cases PasCase UL Slimfast Bonus Choc Royale 6 Bandaid Family Twin pack 12 Sathers Pixy Stix 12 Gemini Video Tape T120 24 House Brand Aspirin 5 Gr 12 House Brand Complete Allergy Caps 24 Act II Micro Butter 144 House Brand Pain Rel Caplets 500MG 24 House Brand Gesic 24 Sathers SF Asst Sour Mix 12 Top 10 items by of customer requests picks SKU Act II Micro Butter Beach Ba Set Act II Micro Lite Butter House Brand Pain Rel Caplets 500MG Act II Micro White Cheddar House Brand Complete Alergy Caps House Brand Ointment Triple Antibio Wrigley PlenTPak Big Red Wrigley PlenTPak Doublemint UL Slimfast Bonus Choc Royal November 21 2007 Top 10 items by of pieces sold Upper Deck Baseball Low Act II Micro Butter Score Baseball Series II 35 348 Act II Micro Lite Butter 148 570 Topps Wax Pack Baseball 26 276 Act II Micro White Cheddar 132 553 Wrigley PlenT Pak Doublemint 77 526 Act II Micro Natural 92 377 Wrigley PlenT Pak Big Red 67 530 Hershey Resse Peanut Butter Op 29 310 Observations Different views tell us different things is Cases moved is of interest to receiving putaway and restocking operations because each case must be handled separately to put it on a shelf is Picks give us a view of which SKUs have the highest labor requirements is Pieces sold give us a view of sales clerk effort notice they spend most of their time ringing up popcorn and baseball cards November 21 2007 Observations continued There can be some surprising issues with ABC analysis For example notice that the highest SKUs in terms of cases moved has very few pieces per case and hence a relatively small number of pieces moved When considering mature products eg staples they tend to not be skewed in the number of picks ie it falls off slowly compared to fashion products eg the top 100 selling music CDs out of over 100000 make up 25 of all sales Understanding and improving warehouse operations If we want to give priority to certain items how do we decide which items go to the priority area Should we have separate areas for pallet picking and case picking How effective will zoning strategies be for our operations Are there ways we can anticipate customer actions November 21 2007 November 21 2007 SKU level pallet vs case vs What if we looked at pallet versus case and case versus broken case profiles in terms of oorders Em What do we learn The distribution of pallet and full case are such that we can separate pallet picking and zone picking operations and not pay a large penalty for mixed orders iE39 warehouse within a warehouse The same is not true when comparing full and broken case The cost will probably be to high to separate these operations November 21 2007 Distribution of order mix of families Suppose we support 3 families of SKUs F1 F2 and F3 We observe the following percentage of total orders e1 F1 only occurs 25 of the time e1 F2 only occurs 30 of the time e1 F3 only occurs 20 of the time e1 F1 and F2 occur 2 of the time e1 F1 and F3 occur 5 of the time e1 F2 and F3 occur 13 of the time EU F1 and F2 and F3 occur 5 of the time What do we learn Zoning by family will be a very effective strategy F2 and F3 should be located close to each other Related issue family specification E by item type E by item form very useful to have family types with same material handling requirements e1 Can family redesign lead to useful patterns November 21 2007 Correlation In many cases if one item is ordered then there is often times a complementary good ordered How do we profile Simple techniques FE Excel spreadsheets FE Access databases Software solutions FE Data mining software 20 November 21 2007 Overview I What is fastpick I Cost tradeoffs I Design Issues I Example I Slotting complications I Managerial Insights 21 Fastpick Area The fastpick area can be considered as a warehouse within a warehousequot note it is often also called the forward area It allows picking operations to be concentrated in a small area thus lE39J reducing pick costs lE39J increasing responsiveness to customer demand November 21 2007 22 Design Issues 1 What SKUs should we put in the fastpick area 2 For those SKUs stored in fastpick how much volume should be allocated to them 3 How big should the fastpick area be Cost Tradeoffs Each item in fastpick requires an additional restock from storage compared to if that item were only picked from the storage area Any time an item is picked from fastpick the cost is less than if that item were picked from storage lE39 smaller area and hence travel distance El 80 of picking time is spent on travel Tradeoff El Extra cost of restocking vs lower cost of picking November 21 2007 23 Estimating Restocks The fastpick area is maintained by restocking One way to estimate restocks is to approximate product by fluid flow In this case let e1 is cubic feet per year that SKU fows through the warehouse e1 vis volume cubic feet of SKU istored in fastpick The number of restocks per year then is V l November 21 2007 24 November 21 2007 2 Allocation of Fastpick Volume If we use some special mathematical techniques we can show that the optimal space value for each SKU in fastpick is given by V i Other Possible Policies Which of these two policies is best e1 Equal Space Allocation assign each SKU in fastpick the same amount of space E Equal Time Allocation assign each SKU in fastpick an equal time supply 25 The fraction of available storage that should be devoted to an SKU is J7 Each storage bay ie shelf section should be restocked at the same rate Example I 2 SKUS Aand B fA16 fB1 V1 Two heuristics vA vB Total restocks iEl Equal space allocation 12 12 32234 El Equal time allocation 1617 117 171734 El Optimal R 108 02 16 1 Restocks A 20 B 5 Total25 08 02 Volume ratio 0434 Restockratio 4 51 Observations November 21 2007 26 Observations Continued The last observation has some important managerial properties lE39 Restocks should be distributed uniformly with no storage areas that are more active than others lE39 If your restockers are restocking some areas more than others ask them then your policy is out of balance More space should be given to those that are restocked more frequently Hot and Cold Spots Allocate more volume to i and j and less to k November 21 2007 27 Observations No difference in other policies and both are bad The more different the SKU rates of flow 6 values are the more important it is to use the optimal allocation It can be shown that equal time and equal space allocations can have as many as twice as many restocks as the optimal policy November 21 2007 28 How to compute the benefit There are two cost components of the benefit savings for each SKU for each of its p picks Cost for each SKU for each of its picks and ll restocks CVi spi cr Vi quotbenefitquot of puttingi into fastpick c cost per restock pl number of picks for itemz39 number of restockings per unit time eg per month v I Minimum Sensible Storage If we are going to put an SKU into fastpick then it shouldn t be too small of an amount The minimum amount to put is given by or SP since this is the value of vthat results in a net benefit of 0 November 21 2007 29 November 21 2007 Problem Formulation Formally we want to choose the SKUs to place in fastpick in order to maximize this benefit n maxzcxvi i1 st iv SV i1 v120 Solution If we solve the formulation mathematically we find a very important ratio AizpiJZ which we will call viscosity It represents the labor effort required to move a given flow through the warehouse It turns out the the SKUs that are put in fastpick are those with the highest viscosity November 21 2007 Procedure for deciding which items go into fastpick Sort all SKUs in decreasing order of viscosity Successively evaluate the total net benefit of the following strategies El no SKUs in fastpick El rst SKU in fastpick El rst two SKUs in fastpick El El Continue until the benefit starts decreasing The strategy that maximizes the benefit is best November 21 2007 Fastpick Size Can show that the optimal size Iof the fastpick area is given by
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'