When I was contacted for an article on supply chain management, for publication towards the end of 2002, two topics came to mind. One of them was a good one to begin a year with, and could form the basis of a “new year resolution” article. The other topic forms the basis of this article.

Just as you would introspect about and evaluate a number of other things at the end of the “old” year, how about evaluating your supply chain? Admittedly, it takes a little more time than the six days between Christmas and New Year, but we can certainly try to lay the foundation here. Remember, you cannot hope to improve what you cannot measure, and since the buzzword is supply chain improvement / optimisation / effectiveness, we do need to look at the measures as well.

Traditional Measures Will Not Get You There

In their efforts to improve profitability or just to sustain the businesses, most companies face a dichotomy in satisfying each customer’s needs and in keeping costs under control. In this context supply chain management is mainly seen as a means to contain costs. Thus, the traditional key measure many managers apply to effective supply chain management is the cost of their supply chain operations – the lower the cost, the better the supply chain looks to them.

However, even the most hard-nosed manager will acknowledge that it is virtually impossible to do this on a sustained basis – cutting “fat” too deeply can lead you to cutting muscle – similarly profitability, market positioning, competitive advantage can be whittled away if supply chain management only focuses on cutting costs.

Even if you just focus on costs, what costs will best indicate supply chain effectiveness? The cost of inbound and outbound logistics? How about the costs of inventory carried in the various distribution centres? What about work-in-progress? While we are looking at costs, let us not forget the other costs associated with sourcing, and distribution, including manpower costs which do not get covered elsewhere. And finally, if goods hit the market late due to a poor supply chain performance, discounts and markdowns need to be considered as well in the costs incurred by the supply chain. So the thinking that cost is a straightforward measure is, in itself, an incorrect assumption – if you think so, you are probably over-simplifying or under-estimating the costs involved.

Let us then look what other measures might be available. In a previous article I mentioned the need to integrate supply chain management with the company’s business strategy, rather than treating it as a back-office function with dirty fingernails and greasy elbows. In my view effective supply chain management must work backwards from the customer needs in mind. Adopting this approach can enable companies to add financial and business value not only in the long term but sometimes immediately.

Once you look at supply chain management this way, as emanating from customer needs and being integrated into every other function of the business, you begin to realise that there needs to be another way to measure its success as well as taking the key decisions related to supply chains: location, production, inventory and transportation .

Time and space will not permit me to detail the various methodologies, but I believe it is worthwhile highlighting one as the most comprehensive, if not complete, method. Even within this there are several detailed layers, which can only be briefly touched upon here.

“From Your Supplier’s Supplier to Your Customer’s Customer”

The Supply-Chain Council’s Supply Chain Operations Reference (SCOR) model is a method of benchmarking and measuring improvements in supply chain performance. The Supply-Chain Council was formed in 1996-1997 as a grassroots initiative by individuals representing companies including AMR Research, Bayer, Compaq Computer, Pittiglio Rabin Todd & McGrath (PRTM), Procter & Gamble, Lockheed Martin, Nortel, Rockwell Semiconductor, Texas Instruments etc.

SCOR, now in its fifth version, is a cross-industry reference model that contains standard process definitions, standard terminology, standard metrics, supply-chain best practices, and enabling information technology. The SCOR model defines common supply chain management process, and matches them against “best practices”. The model was designed to enable companies to communicate, compare and learn from competitors and companies both within and outside of their industry.

SCOR includes all customer interactions from order entry through paid invoice, all product transactions (whether physical or service) and all market interactions from understanding demand to fulfilling it at each individual order level.

The model works primarily with a three-level pyramid.

Level 1, the Top or Process-Type Level, defines the various process types and performance targets at the enterprise or entity level. At this level, the company is essentially defining its competitive position and operations strategy. This includes its competitive performance requirements, performance metrics, its supply chain scorecard and gap analysis, and a project plan.

This level highlights five distinct management processes: Plan, Source, Make, Deliver, Return. (For reasons why “return” is now being added to the supply chain processes, please see the article on reverse supply chains.)

• Plan: This process includes the assessment of supply resources, aggregate and prioritise demand, plan inventory, distribution requirements, production, material and rough-cut capacity of all products and all channels, make-or-buy decisions, as well as product cycles.
• Source: Sourcing infrastructure and processes include supplier evaluation, certification and feedback, quality monitoring, negotiation and vendor contracts, as well as processes dealing with the receiving of material.
• Make: This concerns production, execution and managing “make” infrastructure, including manufacturing, testing, packaging, holding and releasing of product are undertaken here.
• Deliver: This comprises order management (including customer interaction from raising quotations through entering orders), warehouse management and transportation management. This also includes creating and maintaining customer databases, product and price database, and credit management during the customer interaction, as well as channel management rules, order management rules and managing delivery inventories and managing delivery quality.

This level goes into the next layer of detail. For instance, under the Planning Process, at this level Plan would include Plan supply chain, Plan source, Make to stock, Deliver make-to-order etc. At this level, as the company breaks its processes down, it can uncover process inefficiencies and can even move towards flattening the chain. Level 2 allows a degree of “what-if” analysis and therefore an evaluation of the impact of potential improvements.

Focusing on the material flow, this level includes the geographical and thread diagram for the as-is and the to-be process.

Level 3, the Process Element Level, provides the information required for successfully planning and setting goals for supply-chain improvements. It defines the process elements and includes inputs and outputs, performance metrics, best practices where applicable, and system capabilities needed to support best practices. This level specifically gets into information and workflow analysis, and helps to align performance levels, practices, and systems. This level leads to the fine-tuning of the company’s operational strategy. At this level and below, the impact of improvements can be validated.

Level 4 onwards, the Implementation Levels, are company-specific, and includes the organisation structure and people needs, the process and the technology. It focuses on implementation, i.e. putting specific supply-chain improvements into action. These are not defined within an industry standard model, as implementation would be unique to each company.

Process, Not Functional References

The SCOR model is a process reference model rather than a functional reference model.

Thus, it opens out to analysis those processes that involve cross-functional activity – for instance, the Plan process would involve sales & marketing, manufacturing, finance, and logistics among others. It can effectively draw attention to the gaps in the process rather pointing to specific departmental functioning. This in turn can help the company in communicating clearly, without ambiguity and help in measuring, managing, and refining particular process elements.

It helps companies capture the "as-is" state of a process with the objective to achieve the desired "to-be" future state. It also allows the organisation to quantify the operational performance, and set improvement targets based on best practices in similar companies.

The metrics can include a wide variety of performance measures such as delivery performance (delivery in-full, on-time, in-specification is a comprehensive measure of this), order fulfilment performance, fill rate (for make-to-stock), order fulfilment lead time or supply-chain response time, production flexibility, total costs or realised margin, warranty costs or returns processing costs (for reverse supply chains), cycle time for cash-to-cash (measure of effective capital deployment), etc.

It is virtually impossible for a company to meet best practice norms in all the metrics. Therefore, the metrics that a company picks should reflect its customer needs and its market realities, rather than a “do-all, be-all” approach. Some examples are shown in the following table.

- Increasing Deliveries-in-full On-Time (DIFOT) from existing levels of 20% to improved levels of 95% (combining the metric of “on-time delivery” and 100% “fill-rate”)
- Increasing stock-turns by 30%
- Reducing supply lead time from 16 weeks to 11 weeks

It identified specific operational complexities that stood in the way, including:

- Large number of product SKUs (15,000-20,000)
- Multiple worksteps (30+) of which only 20-30% were being done within the company leading to enormous effort and time being spent in monitoring suppliers
- Overload of unstructured information (e.g. faxes, emails)

Its structured steps to improve supply chain performance were driven by efforts to:

- Increase visibility across the chain
- Synchronise planning and workflow across functions and companies
- Inter-related response between activities
- Fixing task accountability, while pushing collaboration

In 2001 Pacific Brands was sold for A$730 million to private equity group CVC Asia Pacific and its consortium partner, Catalyst Investment Managers. Its expected sale price the next time around is expected to be over A$1 billion. While the higher value is certainly not driven entirely by higher supply chain efficiencies, surely an improved supply chain has played a significant role.

Potential Gaps in the SCOR Model

Some of the areas that SCOR explicitly excludes include sales and marketing (demand generation), research and technology development, product development and some elements of post-delivery customer support. All of these have some impact and influence on supply chains, and may be brought into the fold as the modelling evolves further.

One major omission in the earlier models, which is now acknowledged in version 5 of the model, is the collaborative nature of relationships in the supply chain. Companies have never competed solely on the basis of their own competencies, but have been dependent on their business partners. Asian (including traditional Indian) management practices have long been based on nurturing chains of relationships to compete more effectively.

The dependencies between companies have recently been highlighted more and more in western management fora as well, with statements such as “the future lies in competing supply chains rather than competing companies”.

Key examples of the deep impact of such collaboration include that of Unipart of UK in the late-1980s and through the 1990s. When the British government privatised the automotive conglomerate British Leyland, the better performing portions of the group such as the businesses manufacturing cars, trucks and buses were bought by corporate buyers. A leftover group of parts businesses were taken private through a management buyout.

From that less-than-inspirational beginning, Unipart grew to being a profitable business with sales of over a billion pounds, and now operations in multiple countries. Unipart views the supply chain stretching ‘from raw material to the end-user,’ including suppliers and customers. For Unipart, supply chain management needs a two-way flow of information based on trust and common goals. It advocates the term Supply Network Relationships (SNR) including a pro-active approach, close relationships and continuous improvement. Such collaboration has led to groups being formed of executives from Unipart and a supplier organisation, to get the most effective and efficient results from the supply chain. Sometimes, collaboration can even result in helping to avoid business closure, like Unipart’s battery supplier Tungstone in 1989. Joint teams from Unipart and Tungstone analysed the entire production-distribution chain across the two companies, seeing them as a continuum rather than separate stages. Working together, within two years, they improved business performance, with results such as doubled on-time deliveries from 48% of total shipments to 96% of total shipments.

Adapting to this aspect of supply chain collaboration where different steps in a single process can be handled by different business partners, the latest version of the SCOR model, adds a Level 0 where the company identifies the value network and the points of collaboration along a global value network. SCOR documentation mentions Level 0 as recognising “the uniqueness of Distributed Heterogeneous Processes that comprise the network” (although adding that this is “not in scope”). It also makes a further qualification about the network itself, where collaboration takes place amongst “entities” which may be consortiums, enterprises, divisions or corporate functions – this is clearly a quantum jump ahead from thinking of business processes as within the four walls of one company.

This leads us to one other issue: that of product development. Product development, especially in short life cycle industries (such as fashion) has always been collaborative and across companies. The high degree of “product decay” that has been present in short life cycle industries is now beginning to occur in most other sectors. Computers, consumer electronics, even automobiles, are getting outmoded faster than ever.

In this ever-quicker marketplace, to create a super-responsive supply chain, companies need to adopt a 3-dimensional concurrent engineering approach. 3-DCE involves simultaneously designing the product and its specifications, the associated manufacturing processes and the supply chain (including identifying suppliers, the manufacturing locations, transportation and inventory needs). Traditional approaches begin with designing the product, and then either designing the supply chain or designing the manufacturing process – in the bargain, much time is lost in iterations, re-design and re-specification. The 3DCE approach melds the three into a seamless whole.

An example from the fashion industry : in August a typical retailer or brand would start looking at fashion trends, and start designing a look for the next year’s Summer season. Information and inspiration comes from forecasting agencies, trade shows, and various other places. Over a period of 3-5 months they develop the ideas into physical samples. These are also simultaneously costed. Sales budgets and stock plans are developed based on what is going on in the business right then (roughly one-year ahead of the targeted style). At various times during this “seasonal” process, there are decision-making meetings, where styles are accepted, rejected or changed, pricing and margin decisions taken and orders finalised. Since multiple decision factors and approvals are involved there are several meetings where a buyer / merchandiser, a designer, a technologist, a sourcing specialist and others may get involved together. No doubt, many calendars and travel schedules have to be synchronised for this to happen smoothly. Based on a host of factors, the orders might then be placed with vendors in one or more countries around the world. Typically vendors may take a few weeks to two months to procure fabrics, have them approved by the retailer, and then produce a number of samples, and only once all approvals are finished, put the style into production.

From beginning to end, the process of defining a concept to receiving goods in the retail store might take anywhere from 9 to 12 months for a typical retailer. This one-year advance decision-making on what merchandise and how much to stock, is a bit like driving a car at high speed by just looking in the rear view mirror!

“Fast-fashion” company Zara, on the other hand, largely concentrates its forecasting effort on the kind and amount of fabric it will buy. It is a smart hedge – for one, fabric (raw material) mistakes are cheaper than finished goods errors, and secondly, the same fabric could be turned into many different garments. In fact, for an extra degree of flexibility Zara buys semi-processed or un-coloured fabric that it colours up close to the selling season based on the immediate need. This requires a high degree of coordination between different links in the supply chain, as well as various functions within the company. As far as finished garments are concerned, rather than forecasting, it just quickly produces the least amount possible of what is hot with consumers, and moves to the next hot style fast. With that edge, and a super-fast garment design and production process, Zara’s lead time between conceptualising a style / model, to having it in the stores can be as short as 2 weeks, a result of collaborative product development at its best!

Possibly with the addition and evolution of Level 0, the SCOR model will move to address the effectiveness of the supply chain with product development as an integral part, as I believe it should be. Product development is linked to customer needs, existing or potential, and if the supply chain is to be measured, the impact of product development has to be accounted for.

The measurement and improvement of supply chains is no easy feat – but selecting a tool and framework such as SCOR, and diligent application, can bring benefits even earlier than you could anticipate.

© Devangshu Dutta, 2003