NSM Sourcing

Purchase Plan Simulator

When and how much should you buy for a known production or project plan? Enter your case, simulate the 52 weeks week by week, react to events, and compare your plan with the classic methods and the calculated best plan at the end.

A warehouse with half-full shelves

Everything is calculated in your browser. Nothing is sent or stored.

  1. 1 Sector
  2. 2 Case
  3. 3 Simulation
  4. 4 Result

1. Choose your sector

Your sector decides which boxes appear and which example case is loaded. A representative example case is loaded for each sector (no real company is named). You can change every value afterwards.

  • I cannot find my sectorTell us your sector and we will consider adding a profile. On the contact page, write “Purchase Plan Simulator: sector suggestion” as the subject.Contact page

Methods and sources

Lot-for-lot

Orders each week's net requirement one lead time ahead. It holds no stock, but pays the order cost for every need and cannot use the discount tiers.

EOQ

Harris's (1913) formula Q* = √(2·D·K / H), with D the annual requirement, K the order cost and H the unit price × annual holding rate. It orders Q* when the net requirement is about to run out, and follows the minimum order quantity and multiple. It assumes even demand.

Silver-Meal

The heuristic of Silver and Meal (1973): it keeps adding weeks to a lot until the average cost per period (order + holding) starts to rise. It ignores unit price and discount tiers.

Calculated best plan

An extension of the dynamic programme of Wagner and Whitin (1958): an order arriving in week t covers weeks t…j; cost = order cost + price(Q)·Q + holding. Shelf life limits how long a lot can cover (Hsu, 2000); discount tiers set the price by order quantity (Li, Ou and Hsu, 2012). Surplus from the minimum order quantity and multiple is carried as stock. It is recalculated every week with what is known at that moment. That is why it is called the “calculated best plan”: a very good plan, not a proven optimum.

Had you known the events

The same plan as the calculated best plan, but it knows the realised requirement, price rises, technology change, supplier delays and which lot a quality rejection will hit in advance. If one of the plans above turns out cheaper, it takes that one. It is an upper bound; efficiency is the cost of this plan divided by yours, and at most 100.

Common rules

Common rules: holding is charged weekly on the value of stock at the order price; stock in quarantine is held too. The purchase amount is booked in the week of the order. Every method runs on the same event timeline as you, re-planning every week without knowing future events.

Sources

  • Wagner, H. M. and Whitin, T. M. (1958). Dynamic version of the economic lot size model. Management Science, 5(1), 89–96.
  • Silver, E. A. and Meal, H. C. (1973). A heuristic for selecting lot size quantities for the case of a deterministic time-varying demand rate and discrete opportunities for replenishment. Production and Inventory Management, 14(2), 64–74.
  • Harris, F. W. (1913). How many parts to make at once. Factory, The Magazine of Management, 10(2), 135–136, 152.
  • Hsu, V. N. (2000). Dynamic economic lot size model with perishable inventory. Management Science, 46(8), 1159–1169.
  • Li, C.-L., Ou, J. and Hsu, V. N. (2012). Dynamic lot sizing with all-units discount and resales. Naval Research Logistics, 59(3–4), 230–243.
  • APQC. Inventory carrying cost as a percentage of inventory value (open-standards benchmarking measure). www.apqc.org

This is a decision-support and learning tool; results depend on the values entered. The calculated best plan is a very good plan under shelf-life and discount constraints; it may not be the mathematically exact optimum.