Getting excited about ‘new’ tools.
It can be refreshing to learn about new tools and techniques that will make our supply chain lives easier. But, as always, wear your sceptics hat and ask such questions as ‘better for who’ and ‘free for whom’?
I recently read an article about the benefits of implementing ‘multi-echelon inventory optimisation’ (MEIO). This article is one of many written on the topic over the past 15+ years, so the concept is not new; but why has it not been widely adopted over the years? A thought is that few software firms promote an ‘off the shelf’ MEIO application due to the complexities of defining many different types of supply chains and therefore not able to standardise the process. This leaves it to specialist consultants with the ‘know how’ to build a solution and for clients with a belief in the concept, an internal evangelist to sell the proposal to senior management and a finance director with deep pockets!
Inventory optimisation techniques identify the minimum amount of inventory required at locations (nodes of the supply network) to achieve availability targets at the lowest total cost. Ideally, the algorithms will take account of the complex relationships between target service levels, fill-rates, lead times between nodes, costs, capacities, variability (for example forecast error and supplier delivery performance) and risk. This will be for each item in each location of a business, across all the supply chains in its supply network. However, the study may be limited to using only the fill rates at each node and lead-times between nodes for each item.To achieve optimum inventory, the technique is able to factor in the substitution of items and nodes across the supply network.
At this point, your question could be “Does the lack of this technique mean being left behind?”. This is unlikely, because MEIO is only applicable in a few situations. Multi-echelon means multi-level and across your core and extended supply chains, how many levels do you know about and importantly, have influence over? The actual situation is likely to be restricted to your own warehouses, plus knowledge about your tier 1 suppliers and customers. However, you would have to work hard to influence their inventory levels.
Core supply chains and optimisation
While it is recommended that you have knowledge of your supply network, only your core supply chains are really relevant to optimisation. The range of operation decisions required for each node of a core supply chain for a Consumer Packaged Goods (CPG – controlled by ‘use by’ date) or Fast Moving Consumer Goods (FMCG) business are limited. The choice of operation that services distribution inventory is either: make to stock, assemble to order (with or without postponement) or make to order. This will define decisions concerning the form and function of inventory at each node.
So, where would it be good to implement MEIO? It requires a situation where finished items have a complex Bill of Material (BOM), with high cost items containing many common components and where substitution is possible. Items should produced along multiple manufacturing nodes over long supply chains and there are many choices concerning where to hold inventory and the volume. One industry that has been suggested as relevant for MEIO is computer manufacturing and assembly, as there are many interchangeable components and inventory stocking points across a large and distributed supplier base. This may be aided by the high level of outsourcing and contracting, with brand name companies and tier 1 suppliers able to insert control clauses in the supply contracts.
Building supply chain intelligence
While MEIO may have limited applicability, the principle of knowing your supply network is very relevant. With a suitable knowledge base, you are able to ‘optimise’ the inventory that you control or influence; not through the use of mathematical algorithms but through the application of supply chain intelligence. Long supply chains and therefore multiple levels (echelons) of suppliers require that you build an intelligence base of your supply network. You may not know that a level 3 or 4 supplier provides critical materials in the process – speciality steels for instruments, special purpose adhesives or substrates used in packaging to prevent the passage of moisture and odours. These suppliers can stop supply to level 2 or 3 suppliers without you knowing the cause, such as a natural disaster, labour dispute, or financial problems.
Supply chain knowledge provides the core input to risk analysis. And this process will inform you of critical materials and components that may require safety stock to be held, even though the items are ‘dependent items’ in the BOM and in principle should not require safety stocks. Instead of MEIO, the approach to consider for optimising your inventory is Inventory Deployment Optimisation (IDO) and Network Flow Optimisation (NFO). These techniques determine where to store or produce inventory and the flow of inventory from the source of supply to consumption.
These decisions are best made with knowledge of the total supply chain for an item (its core and extended supply chain) and are often made in the context of strategic supply chain planning. Example of the range of data that should be collected to enable decisions concerning inventory location and volumes include:
- By item
- demand profile
- forecast errors
- customer service levels required and achieved per location and customer
- Core supply network nodes that are known and can be analysed (which can also be a variable in the analysis)
- fixed and variable costs by resource and demand and supply channel
- supplier and customer lead times, capacities and location
- supplier service and consistency
- inventory carrying cost
- inventory replenishment policies
- inventory cost and profit margin versus transport costs (for example a cement distributor is likely to have more sites holding inventory to minimise the number of deliveries)
- By Core and Extended supply network node:
- the function, capacity and capability
- production, storage and transport capacities and constraints
- inbound and outbound variability
- location risk profile
This approach emphasises that static inventory policies such as ‘weeks of cover or supply’ are no longer applicable, as they are not responsive to changes in supply chain variability and risk, and service level targets or costs.