Inventory- a stock or store of goods • Dependent demand items- components or sub-assemblies (In a Roland piano, the bench, for example). Forecast is based on # of related finished goods • Independent demand items- finished goods that have their own demand curve (subject to randomness we discussed during forecasting section
Inventory- a stock or store of goods. Dependent demand items- components or sub-assemblies (In a Roland piano, the bench, for example). Forecast is based on # of related finished goods - PowerPoint PPT Presentation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Inventory- a stock or store of goods
• Dependent demand items- components or sub-assemblies (In a Roland piano, the bench, for example). Forecast is based on # of related finished goods
• Independent demand items- finished goods that have their own demand curve (subject to randomness we discussed during forecasting section
Types of inventories- piano example
• Raw materials & parts (e.g. piano keys)
• Work In Process (keyboard assembly)
• Finished Goods (keyboard, stand and bench)
• Replacement items (keyboard cover handle)
• In-transit inventory
Why keep inventory if it costs so much?
• There are times in which the cost of keeping inventory is less than the benefits derived:
• smooth production requirements as seen in Agg. Planning examples
• decouple operations A distribution company wants to keep distributing even if the ship carrying the next shipment is late!
• To meet our stockout goals. Software is quick-decision purchase- many companies have 0% stockout strategies as a result (I.e. opportunity cost = 100%; inventory cost may equal 50%)
• To capitalize on opportunities. If we have excess warehouse and staff capacity, we may save by buying a lot at a great price.
Why keep inventory if it costs so much?
Ordering: quantity & timing Realities in the real world
• Your order quantity may have to be done for political reasons (new product the president is behind- Edirol example)
• We may not be able to affect the timing of orders. Distribution companies usually have to place 3 or even 6 month orders for highly technological products to smooth production planning. So fixed interval models are developed.
Counting Inventory
• Periodic systems count physically at regular intervals and re-order when necessary. Your accounting audit will require this.
• Perpetual systems (that count inventory as it changes in real time and re-ordering when we hit a reorder point) are almost universally used as the cost of computing has decreased.
• Most companies combine use of both.
Adding math models to your tool kit
• What is the lead time of your order (time between submission & receipt)
• What is your holding cost (includes interest, insurance obsolescence, theft, wear, warehousing, etc.)
• What is your ordering cost (including the cost of the transaction and receipt
• What is your Shortage cost (opportunity)
What inventory do we evaluate?
• Pareto principle tells us that 20% of our items will account for 80% of our orders/ supply requests
• So, use the ABC system to classify value Item demand Unit Cost Annual $ value Class
• Can be used to determine number of re-counts in physical counts (e.g. A’s get 3; B’s get 2; C’s get 1)
• Can also be used to determine who does counts (A’s counted by controller, staff & warehouse; B’s by staff & warehouse; C’s by warehouse only)
The Inventory Cycle
Profile of Inventory Level Over Time
Quantityon hand
Q
Receive order
Placeorder
Receive order
Placeorder
Receive order
Lead time
Reorderpoint
Usage rate
Time
So we’ve evaluated the right inventory. Now let’s order.
• EO Q Model minimizes the sum of holding and ordering costs by finding the optimal order quantity.
• Assumptions: 1) one product at a time; 2) we’re confident in our annual demand forecast; 3) demand is even; 4) lead time is constant (management issue); 5) orders received in one delivery; 6) no qty discounts
• 600 is the optimum order quantity account for discounts
We know how much to order… now, when do we reorder?
• ROP: predetermined inventory level of an item at which a reorder is placed.
• Demand (d) and Lead TIME (LT)
• ROP= d*LT
• Example: Monthly demand is 400. Lead Time is two weeks (.50 months). ROP= 400 *.50 =200
• Reorder when inventory level reaches 200.
• This model assumes static d and LT
What if demand or lead time is variable?
• Then we add a safety stock to help us satisfy orders if demand is higher than expected.
• Company policy: What is our service level? It is the number: 1- stock-out risk. “Our service level goal is 95%. In other words, there’s a 95% probability we won’t stock out.
Handling variability, 2
• We assume the variability is characterized by the normal distribution.
• Turn to page 889. The shaded area under the curve represents the probability of us having inventory, given the variability in the average demand or average lead time.
• So let’s say we have a service level goal of 95%. What is the Z score that characterizes 95% of the area under the normal curve?
• About 1.645
When lead time is variable:
• First example: LEAD TIME variable.
• When lead time is variable, ROP= d* avgLT + z*d(LT) where d= demand rate; LT= lead time; LT=std. Dev. Of lead time
• Get the z score (based on your service level goal) from the table as we saw on the last slide based on company’s stockout policy..
• Given: demand during lead time =400/day
• Lead Time = 5 days, =2acceptable stockout risk= 5%
• STEP 1: get your Z score1-.05 = .95 z (.95) =1.65
• STEP 2: plug in400*5 + 1.65*400*2= 3320
• Reorder when inventory = 3320
ROP= d* avgLT + z*d(LT)
If demand rate is variable:
• ROP= avgd* LT + z* sqr.root of LT * (d)
• assume: avg d =1000; d= 14; LT=4; company stockout policy = 10% risk.
• Z score for .90 = 1.28
• 1000*4 + 1.28* 2 * 14= 4000+ 35.84= 4036
• in real world, d is derived by managers keeping careful records to determine it.