• Fixed
• Process
• Product oriented
• Cellular layout

There are many factors conditioning the layout:

• Materials
• Machinery
• Workforce
• Movement
• Waiting times
• Auxiliar services
• Building
• Changes

To design a layout it is also useful to use the SLP methodology, that is outlined here:

• Toys are strongly season products: sales a re concentrated in 3 months
• Toys life cycles are very short
• Manufacturing process needs to be “stock to order” which implies high risks in stocking inventory.
• Since China is its largest supplier and lead times are long, Famosa is dependent on the intensive use of forecasting.

They had some problems to manage their logistics:

• Not structured porudction plan.
• Operational and logistics costs were too high
• High level of capacity with huge semi.fnisihed inventories.
• Products suffere high rates of obsolescence.
• Outsourcing in China used only as a source of components, products were still finished in Spain.
• Production in Spain gave flexibility but incurred high cost inefficiency.
• Distribution carried out from a single point only.
• Dispatch of products packed loose without using pallets.

The redesign of the supply chain was the solution. And new distribution network appeard:

• Implementation of central and regional distribucion centres:
  • Goods are dispatched direct from the DC in China to the market and o the central DC in Spain.
    Some goods are also dispachned form Onil.
  • They have warehousing accros Europe.
• Creation of a consolidation center in China:
  • The objective was to take advantage of logistics and fiscal benefics.
• They adapt logistics processes to the consumer goods sector standards.

It’s important to think that distribution networks never are designed. From time to time they are re-evaluated, like what Famosa did on 2003. Many factors change son distribution networks are usually adapted.

Some years ago I did an study together with other colleages an study about the Spanish Industry of toys. You can download it by clicking in here. You can find below the presentation (it’s in Spanish):

 They base their distribution on the followign principles:

• A global distribution network: Big distribution network
• Large volumes: Are on the manufacturing side.
• Flat packages: This is important part to increase the value to weight ratio
• Low cost: compared with traditional manufacturers.

On this video from Discovery Channel they explain how IKEA does thigns:

On the subject “Innovation” we’ve also worked on innovation on the IKEA supply chain. This could be a diagram of the IKEA supply chain:

Here you can find the slides she used:

But to do that there are different options:

• A spoke-hub distribution paradigm:”a system of connections arranged like a chariot wheel, in which all traffic moves along spokes connected to the hub at the center. The model is commonly used in industry, in particular in transport, telecommunications and freight, as well as in distributed computing”. Example: TNT
• Web connections: Used by DHL and connects point to point destinations without a central distribution center.

There are important aspects related with demand:

• Seasonal demand: Like on the toys
• Life cycle phase.
• Life limited products.
• Predictability level of demand: Thre are different methods of forecasting.

The furniture is a product with a value to volume ratio terribly low and transporting its really expensive. Then suppliers use to install close to their customers. If you manage to increase the value to volume ratio you will be able to design centralized warehousing.

On this product there are full of empty spaces, so it was really expensive to transport (this is not square):

But the new one is different and it is possible to disassemble the hypopotamus. So transportation costs reduce a lot.

CHEP is the owner of every pallet. You save a lot of money of fuel because you don’t have to transport empty pallet. Instead of moving pallets they save the moving empty pallets. They are using RFID to check where the pallets are.

Here there is an interesting video from CHEP explaining how they are using radiofrequency in their own pallets:

• People with people
• People with macines
• Machines with machines
• Products with products
• Products with machines

The point is that barcodes permits communication between products and people, but RFID does not, they need additional information that can be read by a person.

Fields of application:

• Obcject tracking (vehicle tracking): where are they, are they moving…
• Health and safety tracking: special important for police or ambulances.
• Fuel and speed management: Can be controlled from a center.
• Sales order transmission
• Route planning
• Driver management
• Vehicle diagnostics

Route planning:

Serves to arrange different transport orders to tours of a vehicle fleet. There are different standard problems like:

• Travelling salesman problem: Find the shortest way of a circular tour (starting point till end point) that is as cost effective as possible that visits a certain amount of customers exactly once.
• Vehicle routing problem: is an extension of the travelling salesman problem which various vehicles are available at a depot.
• Pickup and delivery problem: consignments are picked up at one place and transported to their destination.. Pick up locations and the destinations have to be in the same tour and a full truckload PDP is supposed. It’s similar to what some low cost airlines do like Ryan Air.

In route planning there are 2 approaches:

• Dynamic route planning.
• Static route planning.

Location tracking:

Controls supply chains and move products to the market faster. It is also necessary to monitor assets, prevent inventory loss and track vehicle fleets.

Companies must consider how to track inventory an a smaller area or across a wide area.

Related to tracking technologies there are different alternatives:

• RFID: Local area and indoor tracking.
• GPS: Signals received from satellites to track movements of objects moving great distances. A very interesting video about GPS can be found by clicking here.

Advantages:

• Improve efficiency and productivity
• Reduce operating costs
• Speed up logistics activities
• Transparency of all the transport events.
• Automatical data transfer from the order entry system.
• Optimal order distribution to the tours (cost, time and customer optimal).

This is a nice video about RFID:

Transportation is about you won something and give it to a third party on certain conditions. It takes some time and many things can happen during that period of time, for example, a change on price conditions. It’s also interesting to know that it also defines who is the owner of the product during the transport.

There are many different alternatives. It’s main characteristics are:

• Does not cover transfer of ownership and property rights. It’s only about risk.
• With international transaction its compulsory. Otherwise you will not be covered.
• Define key elements of international contracts of sale.
• Reflect the parties’ agreement to divide costs and risks between parties
• Provide the buyer and seller instructions regarding the carriage and delivery of goods.
• Can be used for purchasing parts both within the EU and outside the EU.

The following picture shows a transportation chain from moving the goods from a factory to another market.

So chemistries needed a system to control their inventory and to accurately get provided. Then they decided to use Kanban. It consists on a set of cards related to each product. There was a card at the end of each stock there was a card at the end of each product. They get the cards and send it to the warehouse and move it back to the pharmacy. The card identifies both the medicament and the chemistry. Each product has its own card on the place where it is. With the card you do 2 things:

• You communicate: You give it to your supplier
• You control your inventory: You know when you need to reorder

The chemistry is visited every 4 hours by the supplier, getting the quantity of each product needed (according to the number of cards).

Nowadays the movement of the card is not necessary because of the use of the bar codes and internet. The system automatically knows what the chemistry sells and directly communicates with the warehouse.

According to Ann Campbell from University of Iowa, the objective of routing problems in transportation is “to minimize the distribution costs during the planning period without causing stockouts at any of the customers”.

The most general case can be modelized by the Vehicle routing problem that broadly speaking consists on:

• Service a number of customers.
• Fleet of vehicles
• Looking for an optimal solution

It can be applied to several fields. In logistics, for instance, we could use it on invenvtory routing. This is an interesting case study of inventory routing:

“The Inventory Routing Problem (IRP) is concerned with the repeated distribution of a set of products from several facilities to a set of customers over a given planning horizon. The facilities produce these products at given rates and have ample storage capabilities for the products. The customers consume products at a given rate and have limited storage capabilities. A fleet of vehicles is available at each of the facilities as well as a set of drivers. The objective is to minimize the overall costs during the planning period.”

You can get the full case study including some data sheets by clicking here.

How can we solve a routing problem?

There are different ways to solve a routing problem, like:

• Dynamic programming
• Linear programming
• Genetic algorithms
• Bellman Ford Algorithm: It really computes the shortest path so probably it could not be used on routing because there would be some nodes without service. The video is really interesting.

Using graphs:

Using a graph is a very good way to represent this type of problems. Besides representing it graphically, there are other 2 interesting representations (Introduction to Algorithms - Thomas H. Cormen, Charles E. Leiserson and Ronald L. Rivest):

• As a collection of adjacency lists: It provides a compact way to represent sparse graphs (|E| is much less than |V|^2).
• Adjacency matrix: It’s use when the graph is dense (|E| is close to |V|^2).

There is open source software to represent graphs. I like quite a lot one called Grafos. Regardless the webpage is in Spanish,  it’s quite good and simply to use. You can download it by clicking here.

General cases:

Routing vehicles:

Travel salesman problem:

Travel salesman with m routes:

Specific cases of the routing problem:

There are other interesting specific cases, like when you don’t know with certainty when planning customers loads demands. That one is called vehicle routing problem with stochastic demand. Here you can find some articles about this topic:

• A vehicle routing problem with stochastic demand – Dimitris J. Bertsimas – MIT: Here Prof. Bertsimas explains how to deal with this kind of problems. It’s interesting because by reading his webpage (research section) he says that his goal is “to propose a tractable theory for optimization under uncertainty”

• Planning approximations to the average length of vehicle routing problems with time window constraints - Portland State University: Dr. Miguel Andres Figliozzi studies the average length of vehicle routing problems with time window, route duration and capacity constraints.
• Vehicle routing problem with time windows and a limited number of vehicles – Hoong Chuing Lau, Melvyn Sim and Kwong Meng Teo

Free time window:

In routing problems it’s interesting the concept of  ”Free time window”. It can be defined as a time interval in which the resource load is less than the capacity. The interval should be at least as long as the minimum travel time

Other interesting stuff:

You can also use some specifi software to study this kind of problems. Particularly, Knime is a good open source software that could be useful to modelize this type of problems. I’m still looking how to do it.

You can find here some other interesting slides related with multi stage routing. It shows some interesting applications to airports or traffic control.

International Business. Environments and operations  - 12th Edition (p. 462) – Daniels, Radebaugh and Sullivan

Download from here the excel spreadsheet

Other languages (French, English and Italian) are available by clicking here

• To determine the optimal inventory policy for the whole system.
• To keep the cumulative production line close to cumulative demand line.

Multistage inventory is also known as Multi-Echelon system, where we have different layers of inventory:

Managing inventories also implies to accept a level of uncertainty at different stages of the planning process. According to Galbraith uncertainty is
the difference between the amount of information required to perform a task and the amount of information already possessed.

There are different algorithms to manage multistage inventories:

• Kacprzyk and Staniewski (1982):
  • Infinite planning horizon.
  • Fuzzy system, with fuzzy inventory levels, inputs and outputs.
  • Algorithm that determines an optimal strategy to determine the reinstatement of existent inventory levels.
• Park (1987):
  • Economic Order Quantity (EOQ) model from a fuzzy set theory perspective.
  • The order and inventory costs are modelled by trapezoidal fuzzy numbers.
  • Rules to transform the fuzzy cost information in precise inputs for the EOQ model.
• Hojati (2004):
  • Probabilistic-parameter EOQ model
  • Fuzzy parameter EOQ model
  • Simulation to compare the results.
• Porter et al. (1995):
  • Genetic algorithm to solve an inventory-production-distribution problem.
  • The objective is to determine optimal stock levels, production quantities, and transportation quantities to minimize total system costs.
• Samanta and Al-Araimi (2001):
  • Based on fuzzy logic for inventory control.
  • Periodic revision of the inventory with a variable order quantity.
  • Combines fuzzy logic with a proportional-integralderivative (PID) control algorithm.
  • Product inventory at the desired level.
  • Demand variations and the dynamics of the production system.

One of the main conclusions is that there is an intrinsic relation between inventory contrl/management and the production system. Different production systems require different inventory systems.

Definitions:

• Inventory inaccuracy: The discrepancy between the actual and the estimate inventory

Keywords:

• Control: Centralized or Decentralized
• Lot size
• Reorder point logic
• Safety stock: Large or small
• Demand pattern
• Ordering pattern
• Inventory cycles

Inventory planning hierarchy

Sources:

• Mula. J., R. Poler, J.P. García-Sabater, and F.C. Lario, 2006, “Models for production planning under uncertainty: A review,” International Journal of Production Economics, Vol. 103, 271–285.
• Premises Inventory control in a distribution environment Information replaces inventory “DRP is simply the application to distribution inventories” MIT Open Course Ware
• Inventory systems in practice – MIT Open Course Ware
• Stanley B. Gershwin, “Single-stage, multiple part type, MTS production”, MIT Open Course Ware
• Stanley B. Gershwin, “Multi-Stage Control and Scheduling”, MIT Open Course Ware

Importance of inventory costs.

What are the typical costs involved? (inventory cost drivers):

• Facility costs
• Human capital: The cost of labor to maintain storage
• Finance costs.
• Management costs.
• Procurement costs: costs of purchasing.
• Inventory turnover.
• Sock accuracy
• Service levels: critical goods, non critical goods, scheduled delivery

Strategies to better manage inventory:

• Vendor managed inventory: lower cost of supply
• Postponement logistics: process of postponing service of goods until demand, offering a faster respond to a designated market

But could we explain what is inventory control in 2 minutes?

By managing inventory we should be able to answer the following questions:

• How much to purchase?
• How much to manufacture?
• When to buy?
• How often to buy?

By controlling inventory we should work on

• Reconcilying: Difference between how many units do we have in our warehouse vs how many units the IT system belives
• Inventory turnover: Annual sales/Stock
• Days of stock your are holding: Stock/Daily sales

Concepts:

• Oder up to level (OUL)
• Economic Order Quantity: Depends on the square root

Important parameters:

• Review period
• Lead time
• Stock level

Reording policies:

• Reorder point (continuous review policy) depends on the product. Usually   if stock<ROP then order (q). It’s the demand during the lead time. ROP=d*L+Z*SigmaSQR(L)
• Periodic review method: if (tiem has arrived) then order (OUL-Stock). It’s quite used, specially when you manage many products. You need a calendar, because each product has a different review time. Here the quantity we order is not constant

Both policies could be considered only one (s,S): When time arrives, if the stock is less than S, then order (S-stock)

Time and stock is almost the same thing

Order up to level means the maximum expected demand during (review period + lead time)

From time to time I look at my stock. Once I check it I order for a different quantity. I have to control the time, the stock and the quantityDepending on my stock level I order

Interesting links:

• Influence of a truck strike over Spanish production system:http://www.guardian.co.uk/world/2008/jun/09/spain

The whole Mercadona stock is only 2 days

Price:

The amount of money charged for a product or service, or the sum of the values that consumers exchange for the benefits of having or using the product or service.

Click on the image to enlarge

Cost:

Costs set the floor for the price that the company can charge for its product. The company wants to charge a price that both covers all its costs for producing, distributing and selling the product, and delivers a fair rate of return for its effort and risk. There are different classifications of costs. The more general one is the one that defines fiexed costs and variable costs. Fixe costs do not vary with production or sales level while variable costs vary directly with the levelo of production.

Value:

Buyers’ perceptions of a product value.

Defining prices:

Usually a combination of the 3 previous concepts is used to determine the price of a product. This concept should consider:

• Product costs.
• Competitors’ price and other external and internal factors.
• Consumer perceptions of value.

According to that we can define 2 main pricing strategies:

• Cost based pricing: Product driven process. The company designs what it considers to be a good product, totals de costs of making the product and sets a price that covers cots plus a target profit.
• Value-based pricing: Customer driven process. The company sets its target price based on customer perceptions of the product value. The targeted value and price then drive decisions about product design and what costs can be incurred. As a result pricing begins with analysing consumer needs and value perceptions and a price is set to match consumers’ perceived value

Soutvr: Philip Kotler – Principles of Marketing, 3rd European Edition. p. 587

On the next video, we can see an interesting lecture from Philip Kotler at the London Business Forum where he states that:

Our job is to create, communicate and deliver value to a target market at a profit.

Christopher, M. (1998), Logistics & Supply Chain Management, 2nd ed., Financial Times/Prentice-Hall, London

From Supply chain came Supply chain management:

The task of integrating organziational units along a suplluy chain and coordinating material, infomration and finacnial flows in order to fulfill customer demands with the aim of improving the competitiveness of a suppply hain as a whole (Stadtler 2004)

Usually people have problems planning because they usually have no time at all to plan, only to implement. They usually become firemen, thinking only on a one week timescope.

There are two important flows to manage:

• Forward flow: Move things forward
• Reverse flow: Move things back to the company.

The fact of having to manage two flows at the same time makes logistics more complicated.

It’s important to have in mind the time frame. We are talking about 50 years timescope. That means that there can be a lot of different things happening during last 50 years that could explain this big increasement of the GDP. For me one of the most important thing is the stability that we have lived in this region of the world. It m

After the 2nd World War United States was holding nearly all the production in the world. They needed a market to sell their products and Europe was a very attractive market. Everything needed to be rebuilt. After this period came a large period of stability in Europe, meaning that there was security allowing people to start new businesses.

Financial systems have developed quite well creating these sense of security. There financial system that more or less works.

After the Oil crisis in the 70′s, the price of the oil has gone down quite a lot, reducing a cost that is present in the whole economy.

Having infrastructures have been another key point: roads, ports, airports, railways, etc.

Standardization is an important thing that has helped on developing the world. Things have to be done in a specific way. Inside of this topic we include what could be unit load.

Among most all the countries there has been trade agreements that helped the development of the production accros the world.

Something interesting is to see how national governments have influenced in economy by injecting contingent measures during difficult times. On the following graphic we can see how the GDP has evolved during last 28 years and how governments have invested money to interact with GDP.

GDP from 1980 till 2008 and contigency measures

On this video we can see to Prof. Joseph Stiglitz explaining things about the right use of the GDP.

Interesting links:

• Video: Interview by the World Trade Organisation about International trade figures
• Report about a decline of 9% in the global trade
• Comparison of GDP for OECD member countries
• Understanding national accounts (by OECD)

Main concepts:

• Infrastructure
• Unit Load
• Outsourcing
• Open markets
• Stnadardization
• Trade agreements
• Stability