More Kanban Calculations

Kanban Card

First listed various formulations of calculating kanban quantities in July 2006.  Here are a few more …

9.  wmarhel at Elsmar Cove writes …

The formula for calculating the number of kanban cards in a system for a particular product is:

(Daily Demand x (Run Frequency + Lead Time + Safety Time)) / Container Capacity

Where:

Daily Demand = Customer Consumption expressed as # of units
Run Frequency = Frequency which you decide to set-up and produce that item. This is expressed as a unit of time. For a five day work week, running the product every day would equal (1), every third day would equal (3), etc.
Lead Time = Manufacturing lead time (processing time + Set-up time + queue time) + lead time for kanban retrieval expressed as a unit of time.
Safety Time = Allowance for variations in demand and supply, also expressed as a unit of time. Keep as low as possible.
Container Capacity = Number of units per container (# of units in a container is always the same number).

10.  World Class Manufacturing has an on-line Kanban Size Calculator that uses the following formula:

Total Required Inventory (TRI) = Weekly Part Usage * Lead-time * Number of locations for stock
# Kanban = TRI / Container Capacity

11.  Oracle uses

By default, the standard calculation is:

(C – 1) * S = D * L

where:

  • C is the number of kanban cards
  • S is the kanban size
  • D is the average daily demand
  • L is the lead time (in days) to replenish one kanban

 

12.  SAP says …

K = ((RT * AC)/CONT) * (SF + C)

where

  • K          numbers of Kanban
  • CONT  contents per Kanban
  • RT        replenishment lead time per Kanban
  • AC        average consumption per time
  • SF        safety factor
  • C          constant (default 1)

 

 

 

Entropy and the death of peaceful order

Had dinner last night with Richard Zuber, Lean Six Sigma Master Black Belt at Honeywell.  He was a student of mine some years ago at AlliedSignal before the Honeywell merger.  Richard posed a question as to why so many organizations that have mastered demand smoothing and flow eventually revert back to month end madness, chaos and noise.  He’s seen good applications of heijunka and demand segmentation fail.  Why?

 

 

 

Continue reading Entropy and the death of peaceful order    

Logistics Network Modeling

New trade agreements, shifting customer and supplier locations, changing labor and freight rates, fuel costs, new products and markets all conspire to make your current logistic network obsolete.  Gone are the days when a warehouse distribution network would be reconfigured every ten years or so.  Today aggressive supply chain executives keep a constant eye on the architecture.

Before you can optimize you need to get organized and think about all of the variable that contribute to your costs and service levels.  A supply chain logistics model allows network design to minimize inventory carrying, warehousing, transportation costs while meeting customer lead time and on time requirements.  Have too many warehouses?  Where to open a new distribution center to serve an emerging market?  Distribution Center lease coming up for renewal?  Long transit time from low cost country sources worth the carrying and distribution costs? 

You need data, a modeling tool, and a plan.  Six Steps*

 

  Continue reading Logistics Network Modeling