Supply Chain: Everything After Exam 1
Toys R Us: The Biggest Toy Retailers
(Pic) ... The rest is online
Continued
- ex. 50% service level means you can have demand values up to 50% (?)
Normal Distribution
- right = bigger numbers and vice versa
The longer the cycle time...
... the greater the cycle inventory *Need to be familiar with this diagram *Do not confuse dbar=units/period with Dbar=Total demand during lead time Dbar=dbar*L L=lead time DbarL is the shaded part
Possible Scenarios
Constant Demand and Constant Lead Time Variable Demand and Constant Lead Time Variable Demand and Variable Lead Time
FMEA
Failure Modes Effect Analysis
Application Examples
Manufacturing: A manager is responsible for moving a manufacturing operation to a new facility. She wants to be sure the move goes as smoothly as possible and that there are no surprises. Design: A design engineer wants to think of all the possible ways a product that is being designed could fail so that robustness can be built into the product - ex. in an ad, they pitch you being able to drop your phone in a cup of water BECAUSE it's high likelihood that it will happen "A false emergency alert warning of an incoming ballistic missile caused massive panic in Hawaii, as terrified people scrambled to find shelter and prepared for what they thought might be their final moments" - ex. Need to figure out what went wrong
FMEA Objectives
To understand the use of Failure Modes Effect Analysis (FMEA) To learn the steps to developing FMEAs - what problems will I encounter and what should I fix? - A bunch of problems, but what am I gonna work on?
Failure Modes and Effects Analysis
"Failure Modes" means the ways, or modes, in which something might fail. Failures are any errors or defects, especially ones that affect the customer, and can be potential or actual Or, more simply, Ways in which things could go wrong "Effect Analysis" refers to studying the consequences of those failures Failures are prioritized according to how - Serious their consequences are - How frequently they occur and - How easily they can be detected - Effect Analysis is what is the consequence of this failure happening? - Could be a potential failure
Service Level vs Safety Stock
*******Relationship is nonlinear when the service level is close to 1; i.e., need disproportionately high safety stock to achieve very high service level****** --> THIS IS VERY IMPORTANT TO REMEMBER!!!!! - As service level increases, z value increases really quickly (means amount you will need to carry will rise very fast) - closer to one means you disproportionately need more safety stock
In-Class Exercise: A convenience store replenishes its stock of lemonade on a continuous basis. On Monday morning they have 9 cases in stock. Each case contains 24 cans. The reorder point is 4 cases. The demand during the day is shown below. Assume backorders are allowed Replenishment lead time is 48 hrs, after the order is placed. Assume orders are placed at the end of the day. When would they place the next order of 6 cases. (pic on back)
*Need to be very comforatable with how to do these calculations - Don't use excel
Demand for a Product with 3 week lead time
- **Each is one occurrence of a different 3 week period - repetitions where various orders are placed where LT = 3 weeks - not always the same (75/15), it's just a coincidence
ABC Inventory Management
- A tight control because it's a lot of money
SC Project 2
- PDF doc posted with details
Example Continued
- he will tell us whether to round up or down - If it's a 12 pack, and EOQ=10, you would obviously need to round up to 12
Toys R Us: The company's in-stock level on Black Friday was 62% compared to Walmart's 63% and Target's 51% - Is this good or bad? Why?
Bad because for Walmart and Target, Toys are only a section of their store
What is inventory?
Finished goods (end of the spectrum) Work in process (also called work in progress) Transit (things that are transported) Raw materials (beginning of spectrum)
Calculating EOQ
Q/2 is average cycle inventory - 1 cycle is the cycle between 1 order and the next being received
Severity, Occurrence, and Detection
Severity - Importance of the effect (ex. is someone going to die?) Occurrence - Frequency with which a given cause occurs and creates failure modes (obtain from past data if possible or take your best guess) Detection - The ability of the current control scheme to detect (then prevent) a given cause (Think! Can you detect a problem? How?) - ex. would you know if this problem happened?
Inventory Flows
There's 2 charts of this in the slides? Pipe is inventory coming in - water level is inventory level (finished, raw, WIP) - Depends on condition of business
Calculating EOQ again
Total annual cycle-inventory cost: C=(Q/2)*(H)+(D/Q)*(S) - Ex. this is basically just average cycle inventory * H + number of orders you'll place * setup costs or ordering cost per lot where.. C= total annual cycle-inventory cost Q= lot size (in units) H= holding cost per unit per year D= annual demand (in units) S= ordering or setup costs per lot
SC Project 2 Continued
- Absolute answer should be within a single range - Pretty much only diagram is network diagram - Need to make it ONE SHEET on the excel doc/ only one tab!! - Make sure PDF converts correctly - due nov 14th - Make it easy to follow!
Inventory and Supply Chains: Reasons for large inventory (COLT)
- Customer service (more inventory = better service level) - Ordering and setup costs (more inventory = fewer orders) (ex. one order fee versus many order fees) - Labor and equipment utilization - Transportation cost (less than truckload is more expensive per lb)
Example Above
- End of day means demand has already been accounted for - When IP = 470 and 395, do nothing because it's greater than the ROP - Demand at 260 is less than ROP which triggers the need to order! Before you place order, you're below ROP - 245 triggers second order cuz it's lower than ROP
Distribution of Demand
- If mean demand is 50 and actual demand is 1 st dev more, you're on the right side of the diagram - Ex. if z score is 2.5, you are to the right and if it's -2.5 you are to the left - Z score is a way of normalizing the data - Service level = concerned about demand being higher than mean value (higher service level = far right) - 2 st devs is 95% of data, 3 st devs is 99.7% of data
Inventory and Supply Chains: Reasons for small inventory
- Inventory holding cost (because if this is high, you want small amount of inventory) - Cost of capital (you need to put up the money to buy it) - Storage and handling costs
Inventory Terminology
- Lot size (Q) --> size of batch - Lead time periods (L) --> number of lead time periods, normally just referred to as lead time - Average demand per period (d with a line above it) --> ex. if you have a daily demand of 2, d bar = 2 - Average demand during lead time (D with a line above and subscript L) --> if you have a 7 day lead time and daily demand of 2, DbarL= 14 = dbar*L; also referred to as pipeline inventory - Std deviation of daily demand (σd) - Reorder Point (ROP) --> at what level of inventory do you reorder? - Number of standard deviations corresponding to service level probability (Z) --> distributed demand
Continuous Review System
- Mean demand over 3 week lead time (dbar*L) is just the sum of the 3 during that time!!! (yes, average demand is the sum in this problem because average demand over the entire three weeks is the sum of all of the means) - If standard deviations are the same, use other equation (both work out the same way) --> this will not always be the case!! Only if standard deviation is the same! - L = lead time
Continuous Review System Example Cont
- added annual safety stock holding cost because you now have safety stock (didn't have it before)
Managing Inventory
- all about balancing inventory level with service level
Example Cont again
- always using the same equation for IP
Example Cont
- can plot line for various values of Q - Looking for where the 2 lines intersect/ lowest point and that is the EOQ
Independent vs. Dependent Demand Cont
- ex. can aggregate demand for bikes with same rims
My notes in Independent and Dependent Demand
- ex. demand for one bottle is independent of demand for a different bottle - ex. demand of purple bottle is dependent of demand of same bottle overall MRP= if you buy one of these, how many subassemblies do you need? Ex. if you are making a box, you need a bill of materials (BOM 4A - 2 months 3B - 4 weeks 1C - 2 weeks - different lead times - Need to order 4A 2 months in advance, then 3B, then 1C so you have it by December 1st when you need to produce a box
Stages of Inventory Cont (within the SC)
- some raw, WIP and finished - Supplier also has inventory - overall, you can have variants of these things
Independent vs. Dependent Demand
- subassemblies dependent on how many will be sold - ex. bicycle on next page
Periodic Review System: Advantages of the P System
1. It is convenient because replenishments are made at fixed intervals 2. Orders for multiple items from the same supplier can be combined into a single purchase order (ex. if buying from a single supplier you can combine orders) 3. The inventory position needs to be known only when a review is made (not continuously) --> biggest advantage - continuous vs fixed intervals
Stages of Inventory
1. Raw Materials - Materials and parts waiting for processing and/or assembly 2. Work-in-process - Partially finished components and subassemblies 3. Finished goods - Final product ready for sale or shipment
Variable Demand/ Constant Lead Time
A distribution center (DC) in Wisconsin stocks Sony plasma TV sets. The center receives its inventory from a mega warehouse in Kansas with a lead time (L) of 5 days The DC uses a reorder point (R) of 300 sets and a fixed-order quantity (Q) of 250 sets Current on-hand inventory at the end of Day 1 is 400 sets. There are no scheduled receipts (SR) and no backorders (BO) All demands and receipts occur at the end of the day Determine when to order using the Q system
Inventory Management: What is inventory?
A stock of materials used to satisfy customer demand or to support the production of services or goods ex. Chipotle has inventory of different materials
Calculating EOQ Cont
Annual holding cost Annual holding cost = (Avg. cycle inventory)*(Unit holding cost) Annual Ordering Cost Annual ordering cost = (Number of orders/Year)*(Ordering or setup costs) Total annual cycle inventory cost Total costs = Annual holding costs + annual ordering or setup cost
Calculating EOQ Continued
As lot size increases, Q/2 increases linearly - As lot size increases, number of orders decrease - Lowest point is EOQ
In-Class Exercise
Calculate the replenishment lot size Q (assume no safety stock is needed) Calculate the cycle inventory and pipeline inventory Use the data file In-Class ABC analysis and calculate the number of SKUs that are Class A and the number it units in the Class A items 1. Quantity*Value 2. Sort that 2. Sum all quantity*value numbers 4. Divide individual (quantity*value) by total 5. Add and see where you get to 80% Always assume 7 days per week, 365 days per year, and 52 weeks per year unless it says otherwise
Inventory Placement
Centralized Placement (one location) - Keeping all the inventory of a product at a single location Inventory Pooling (too many locations and pooling them down) - Reduce number of inventory locations. Merge variable demands from customers Forward placement - Locating stock closer to customers at a warehouse or distribution center - from where to where? - Well, where are you holding the inventory?
ABC Item Classification
Class A Items: 10-20% of items which account for approximately 60-80% of annual $ usage (could be looking at $ usage, $ value, etc) - Very tight control and tracking Class B Items: 20-40% of items which account for approximately 20-30% of annual $ usage - Tight control and moderate tracking Class C Items: Remaining SKU's (40-60%) which account for remaining 5-15% of annual $ usage (don't focus on C items, focus on items where the money is coming from! A items) - Loose control limited tracking *Need to know rough percentages for exam! - Ex. "we have an inventory problem" .. well 80% of your value probably comes from 20% of your inventory
Types of Inventory
Cycle Stock (Average amount you have before you get the next shipment - ex. you order a batch, use it, then order another batch) - Inventory resulting from batch (rather than unit) ordering or production Safety (Buffer) Stock (what you carry to protect against variations in demand (or in supply) - ex. when you open your doors, there's some uncertainty in demand (might be 2 things normally, then another day it could be 15)) - Buffer against uncertain demand Pipeline (WIP) Stock (How much inventory you have in your facility while waiting for next shipment) - Goods in transit and in-between stages of production Anticipation Stock (build up stock in anticipation of demand ) - Accumulation in anticipation of peak demand Triangle diagram thing --> top is quantity, the next triangle is another shipment the next day - in the basic diagram, by the end of day, Q is 0
Managerial Insights from the EOQ
Demand --> increase in lot size is in proportion to the square root of D Order/Setup Costs --> Weeks of supply decreases and inventory turnover increases because the lot size decreases Holding Costs --> larger lots are justified when holding costs decrease Look at pic cuz not everything is right here! - As demand increases, EOQ will increase by the square root of D - If set up costs decrease, EOQ will decrease - ex. cost of prime account --> the more you use it, the average setup costs are reduced (shipping cost cuz you get free shipping with the prime membership) - Holding cost decrease, EOQ increases
Inventory Models
Demand: Constant, deterministic, stochastic Lead time: Constant, ">0", stochastic Horizon: Single period, finite, infinite Products: One product, multiple products Capacity: Order/inventory limits, no limits Service: Meet all demand, shortages allowed EOQ New Vendor Continuous Review Periodic Review - if conditions are all red, choose EOQ - If purple, use news vendors (but we dont cover it in this class) - ex. single period because most don't want yesterday's news - Green = continuous review and periodic review - Periodic review is like looking at inventory every 6 weeks
Continuous Review System
Determining the Reorder point with variable demand and constant lead time Step 1: Calculate demand over the lead time dbar = Average demand per week (or day or month) L = constant lead time in weeks (or days or months) Step 2: Calculation of safety stock - Choose an appropriate service-level policy - Determine the distribution of demand during lead time - Determine the safety stock ROP = dbar*L+safety stock - understand what demand during lead time is (same as demand over lead time) - You could have higher than expected demand (think about graph) - Difference between this is your SS (average demand and maximum demand/ expected demand and actual demand)
Calculating EOQ cont
EOQ Formula (pic) Differentiate wrt Q and equate to 0 Time Between Orders (TBO) - make sure to multiply by 12, 52, 365, or whatever the relevant units are!! (depends on what you want the TBO in - days, weeks, or months)
Inventory Management Systems
Establish amount of inventory Monitor inventory levels Determine when to replenish inventories Calculate order or production quantities Types of inventory management systems - Continuous Review (RQ) systems - Periodic Review (PT) systems
Next Session: Inventory Management Systems
Establishing amount of inventory Monitoring inventory levels Determining when to replenish inventories Calculating order or production quantities Types of inventory management systems - Continuous Review (RQ) systems - Periodic Review (PT) systems Read "Why is inventory management important to modern retailers"
Independent Demand
Finished goods, end products purchased by customers Raw material commodity stocks, common to many products Demand rate and pattern determined outside of the firm Demand is independent of demand for other products or items We will focus on Independent Demand Items in this course - finished goods
Example
He will give us z values - Ex. moving 10% drops it by over 50% (looking at numbers) *Average demand is Total demand across the three weeks (Says average but you sum)
Cycle Inventory (Cycle Stock)
Inventory needed between orders or production runs Depends on - The lot size (Q) - Varies directly with the cycle time (elapsed time) between orders The longer the cycle time, the greater is cycle inventory - because you need more stuff! - Be familiar with this chart!!! Average between A and B = (Q+0)/2 - If you had 5 at the end of the period, it'd be (Q+5)/2 - Q/2 is cycle inventory
Example: Constant Demand and Lead Time: The on-hand inventory is only 10 units, and the reorder point R is 110. There are no backorders and one open order for 200 units. Should a new order be placed?
Inventory position = On-hand inventory + Scheduled receipts - Backorders IP = OH + SR - BO IP = OH + SR - BO = 10 + 200 - 0 = 210 R=110 Decision: Place NO new order
Inventory Terminology Again
Lot size (Q) Lead time periods/ lead time (L) Average demand per period (dbar) Average demand during lead time (DbarL) Std deviation of daily demand (σd) Reorder Point (ROP) Number of standard deviations corresponding to service level probability (Z)
Dependent Demand
Materials, parts, and subassemblies used in end products Items traceable to specific end products Demand is dependent on the number of end items manufactured To produce 100 bicycles in March, then need 200 wheels in February and 7200 spokes in January Dependent demand items are controlled using Materials Resource Planning (MRP) systems - bold is what he emphasized in class
Inventory ABC Analysis
Observation: 20% of stock keeping units (SKU) account for 80% of total inventory costs (Pareto principle) (roughly) Idea: Manage most important (costly) inventory items most closely Use: First analysis to undertake when attacking inventories! SKU = stock keeping units
Multi-periods and Stochastic Demand
Overage is no longer a big deal - Leftover inventory can be used in the following periods (unlike that in the single-period case) - Cost of overage is holding cost - Possible economies of scale for fixed ordering cost Shortage is more serious - Performance measure: Service level Service Level alpha = Prob (no stock-out) - Need to hold safety stock to achieve service level (ex. to adjust for higher demand) - Only cost is the holding cost of the excess amount - ex. 1 pallette is $10, 3 pallettes is $20, so you might as well order 3 cuz youll use it eventually (higher holding cost)
Toys R Us: What is the Major issue they faced from stock-outs?
People don't come back after you don't have what they need in stock
Pipeline Inventory
Pipeline inventory = DL (with line) =dL (with line) - Demand during lead time = dbar*L
Safety Stock Inventory
Protects against uncertainties in demand, lead time, and supply changes - place an order when you get down to safety stock - if I have maximum probably demand, you would run out at the end of lead time - If you have normal demand, the amount leftover is safety stock - ROP = time of the level of inventory; time you place an order - 7 day lead time and max demand per day = 20; how much would you have to carry? You need to have 140 on hand when you place an order - BUT... if normal dbar=10 per day, you'll have 70 leftover --> 70 is your safety stock - Safety stock helps with maximum demand and accounts for the fact that demand could vary
Risk Priority Number (RPN)
RPN is the product of the severity, occurrence, and detection scores - RPN is a product of the numbers - Can determine top 5 most important based on how high the RPN is
Distribution of demand over lead time
Reorder point = Average demand during lead time + safety stock - ex. average demand during the 3 weeks is 100 and you want 95% service level so you will be to the right. Value needs to be over 100 and you need to figure out standard deviation - Middle value = total demand during the 3 weeks!! - If demand is above the safety stock plus avg demand, you will stock out - Need to have some amount above so you have a 95% chance you will have enough - amount is z*stdev - diff between mean and what you have is safety stock
Continuous Review System (Q)
Reorder point system (ROP) and fixed order quantity system Tracks inventory position (IP) Includes - Scheduled receipts (SR)/Open orders - On-hand inventory (OH), and - Back orders (BO) Inventory position = On-hand inventory + Scheduled receipts - Backorders IP = OH + SR - BO - Scheduled receipts are what you're expecting to receive - On hand inventory - physically, how much do I have on hand - Backorders - what you've promised to someone else - Inventory position is netting out all of these things
Continuous Review System (Q) Again
Reorder point system (ROP) and fixed order quantity system Tracks inventory position (IP) Includes - Scheduled receipts (SR)/Open orders (ex. what's expected to come in) - On-hand inventory (OH), and - Back orders (BO) Inventory position = On-hand inventory + Scheduled receipts - Backorders IP = OH + SR - BO - Fixed ROP, Fixed quantity, Tracking inventory position
Toys R Us: Based on this statement "...subject to significant fluctuations in sales and inventory turnover during the holiday season," What kind of SC do they need?
Responsive SC because they have fluctuations in demand
Rating Scales Cont
Severity: 1= Not Severe, 10= Very Severe Occurrence 1= Not Likely, 10= Very Likely (or low frequency and high frequency) Detection 1= Easy to Detect, 10= Not easy to Detect - stay consistent
Toys R Us: What was happening to their target market? Who are they?
Target market is the parents of the kids - lots are going online now
In-Class Exercise
The demand for a product is stochastically distributed. The product has a 4 week lead time and the demand and standard deviation of demand for each week is given below - what is the total demand during the lead time? - What is standard deviation of demand during the lead time? - What would be the safety stock required to achieve a 97% service level? - What is the reorder point? - What would the safety stock be for the same service level if the standard deviation of demand were constant at 5?
Periodic Review (P)
The inventory position is reviewed at fixed intervals (P) and the order is placed at the end of the review Same EOQ assumptions apply 1. The Lead Time is constant and known with certainty 2. No constraints are placed on the size of each lot 3. The only two relevant costs are the inventory holding cost and the fixed cost per lot for ordering or setup - Fixed intervals - ex. end of quarter inventory
Economic Order Quantity
The lot size, Q, that minimizes total annual inventory holding and ordering costs Five Assumptions: 1. Demand rate is constant and known with certainty 2. The Lead Time is constant and known with certainty 3. No constraints are placed on the size of each lot (ex. no limit on what you can order) 4. The only two relevant costs are the inventory holding cost and the fixed cost per lot for ordering or setup 5. Decisions for one item can be made independently of decisions for other items
Inventory Management: Purpose of inventory management?
The planning and controlling of inventories in order to meet the competitive priorities of the organization - trying to manage different competing priorities - ex. build up inventory for holiday season --> priority is different
Continuous Review System: Advantages of the Q System (FFS
The review frequency of each SKU may be individualized Fixed lot sizes if large enough can results in quantity discounts The system only requires levels of safety stock for the amount of uncertainty in demands during the lead time, which can be lower - during the LT period, what is the uncertainty in the amount I will need?
Rating Scales
There are a wide variety of scoring approaches Two types of scales are 1-5 or 1-10 The 1-5 scale makes it easier for the teams to decide on scores The 1-10 scale may allow for better precision in estimates and a wide variation in scores (most common) NOTE: I use the following options for ratings: 1, 3, 7, 10. Why? - Why avoid 5? It forces people to make a decision and go one way or another
Toys R Us - Sales
Toys R Us went out of business - red line is growth rate --> in 2014, growth rate just got a little less bad - could see growth rate and sales were bad for a while
Economic Order Quantity Cont
Use the EOQ - Make-to-stock strategy with relatively stable demand - Carrying and setup costs are known and relatively stable Don't use the EOQ - Make-to-order strategy - Order size is constrained Modify the EOQ - Quantity discounts - Replenishment not instantaneous (ex. it arrives and is instantaneously available versus needing to do something to it before you can use it)
Anticipation Inventory
Used to absorb uneven rates of demand or supply when demand far exceeds the ability to produce at that time Predictable, seasonal demand patterns lend themselves well to the use of anticipation inventory - ex. xmas lights demand would be flat then spike in november and december - Means there's no way you can produce enough so you need to produce ahead of it - Ex. pumpkin filling --> you can't produce it all in october and november so you need to produce ahead of time - Keep production steady
Inventory Decisions
What is inventory? Why is it held? Stages of Inventory Purposes of inventory ABC inventory analysis (done most often in inventory management) Inventory management systems Independent vs dependent demand systems Why materials management is crucial
Continuous Review System (Q System) Cont
When Demand and Lead Time Are Constant and Certain Reorder Point = Demand over Lead Time - When D and Lead Time are constant and certain, ROP=D/LT - Amount you need on hand is exactly how much you'll consume during that lead time
Continuous Review System (Q) Againn
When demand is Variable and Lead Time is Constant Selecting the Reorder Point When Demand is Variable and Lead Time is Constant -*Demand is variable while Lead Time is constant - What's going on with your on hand inventory?
Toys R Us: "Full and Chunky" What is the risk?
customers are very fical (if you entire shelf is stocked with candylocks and the kids want goldilocks, you're screwed)
Toys R Us: How is Amazon more effectively competing with Toys R Us? Cont
ex. Amazon's price was higher --> New (35) from 22.39 plus 4.99 shipping - Amazon doesn't own the inventory! They just stock it - Amazon doesn't get the risk because if a different toy is popular than they thought, it's fine! - Amazon doesn't pay for inventory --> this is their single biggest advantage - ex. if Toys R Us demand is greater than what they have, they are screwed --> not the case for Amazon since they don't own the inventory
EOQ Example ❖ A museum of natural history opened a gift shop which operates 52 weeks per year. Top-selling SKU is a bird feeder. ❖ Sales are 18 units per week, the supplier charges $60 per unit. ❖ Ordering cost is $45. ❖ Annual holding cost is 25 percent of a feeder's value. ❖ Management chose a 390-unit lot size. ❖ What is the annual cycle-inventory cost of the current policy of using a 390-unit lot size? ❖ Would a lot size of 468 be better? ❖ What is the best lot size for them to order?
the "feeder's value" is H=0.25*60 D= (18 units/week) * (52 weeks/year) = 936 units H= 0.25*60 = $15 per unit When placing orders (and deciding between two alternatives), you care about total cost - look at the case - what do you know and what do you not know? - Other lot size TC = 3600
Why Inventory Matters So Much
this was just a video.. - will need to install a little software that he sends - Too much inventory = dramatically cut prices because expensive to hold inventory
Important things in content heavy slides
usually 2-3 things that are important in the content heavy slides - Upper half of slides and things we talk about