Time to be demanding

LinkedIn +

Today’s supply chains have developed into sophisticated networks that can easily crumble if not enough care is taken. One area where this could prove an issue is in demand management. Lisa Townshend looks at the issues.

In terms of demand management there are two main processes at work: one, the demand is driven based on historical data or forecasts (usually known as a ‘push’ model where goods are moved down the supply chain based on these forecasts – this is the traditional model); two, goods are produced and moved based on actual demand or consumption -a ‘pull’ model, or ‘zero inventory’.

Logistics Manager, April 2016.

This article first appeared in Logistics Manager, April 2016.

For years the push model worked well enough. Now, however, there is a call for supply chains to move operations, where possible, to a demand driven model to be more responsive to customer needs. The rise in e-commerce is a perfect example of why the pull model can be more efficient in these circumstances.

Eric Carter, solutions architect at Indigo Software, says: “Zero inventory stock management is a key strategy for a warehouse to use because it significantly benefits cash flow and actually facilitates new business models. Just about every internet-based retailer has behind them a zero inventory model, whereby they ‘back to back’ the customer order with a PO and ship end product directly to the customer. Amazon’s Marketplace and many other discounted retailing models would be impossible without zero inventory stock management.”

Michael Latimer, vice president sales & marketing at True Commerce, says: “Increasingly retailers and manufacturers are adopting a demand driven supply chain approach to optimise stock logistics and minimise costs. The main challenge that businesses looking to implement a demand driven strategy face is the availability of live, or near live, data from trading partners.

“With the increasing pressure on supply chains to keep processes as lean as possible in terms of matching supply and demand, it can mean that demand may exceed stock availability occasionally, leaving providers unable to stock up, manufacturers unable to complete the manufacturing process and customers without the items they wanted. However, this can be overcome with high quality, real time data that directly links stock availability with demand management.

“To ensure processes are optimised, any data insights should be directly linked to multichannel offerings so that when an item has sold out, all channels, including the web site, are updated instantly so customers aren’t left disappointed.”

Of course the demand driven model won’t work for every sector. However, some sectors are seeing benefits using demand-driven supply chains. Carol Ptak, co-founder of the Demand Driven Institute, highlights the increase in interest: “There has been a shift in the last year from the industry not understanding the inherent flaw of formal planning to people directly interested in Demand Driven Material Requirements Planning (DDMRP) as the solution.

“Globally, DDMRP is being implemented at an accelerating rate from the retail level and through their integrated supply chains because companies achieve dramatic results quickly. MIC, a clothing manufacturer in Colombia, went from financial crisis to being named the best supplier for El Exito in just 18 months – by developing a demand driven supply chain capable of reliably delivering twice the volume with half the inventory.”

Lionel Albert, supply chain cloud director at Oracle EMEA, believes that the difference in acceptance has historically depended on which end of the supply chain you are at. “Manufacturers are already on the ‘pull model’ as they implement demand management and have clear view of the specific needs and requirements of their customers. Retailers on the other hand are still, for the most part, using a ‘push model’, as they need to ensure that their stores have plenty of choice for customers to browse and allow for discretionary purchases that may not appear on shopping lists.

“This results in a major battle between the manufacturer and the retailer to agree quantities, as the manufacturer is one step removed from the end customer and doesn’t have the same level of visibility into what is going on at the point of sale.”

John Perry, managing director at UK logistics and supply chain consultancy SCALA, is surprised at just how quickly the UK is accepting this modern approach: “Supply chain leaders experience at first hand the problems that arise when forecasts are used to drive operations, but I have still been amazed at the depth of interest in the Demand Driven approach. There is a very receptive British audience for this new way of working because new methods are needed to cope with today’s commercial challenges.”

Henry Canitz, product director at Logility takes a different view: “Of all factors that affect business success over the long term, most executives agree that generating accurate forecasts is right at the top of the list and therefore most retailers and manufacturers create some type of forward looking forecast of demand. Accurate forecasts have a significant impact on an organisation’s ability to satisfy customers and manage resources cost effectively. You have to remember, a forecast is not simply the projection of future business: it is a request for product and resources that ultimately affects almost every decision a company makes. That means decisions by Sales, Finance, Production, Logistics, and Marketing are all enhanced or diminished by the quality of the company’s forecasts.”

There are many obstacles to adopting a demand driven model, not least the use of the right software for the job. Canitz adds: “For many companies the initial challenge is building the right foundation of people, process and technology. Excel continues to be one of the primary tools for their forecasting ‘system’ even though study after study has concluded spreadsheets provide inadequate functionality and are riddled with errors. Conversely, research shows that implementing an advanced statistical forecasting solution is a fundamental step to improving forecast accuracy. Spreadsheets and ERP systems simply cannot support advanced planning activities such as ‘What-If’ Scenario Analysis, Collaboration, & Product Lifecycle Planning.”

For Ptak, it’s the mind-set and people’s perception that can be the biggest hurdle. “The first step is the most difficult – rebooting the thought-ware. People need to change how they think so metrics can be changed to something that aligns activity to benefit ROI. Current processes are embedded so deeply that many people just do not believe there could be another way.

“Traditional measures of efficiency, utilisation, overhead absorption and OEE (Overall Equipment Effectiveness) do not tie to improved ROI and in fact lead to a reduced ROI. Implementing smart metrics allows a company to achieve results never before thought possible.”


Albert sees the challenge being one of understanding: “The challenge for retailers is to accurately predict the demand for specific products once they hit the shelves, while still ensuring that customers have sufficient choice. With limited floor space to work with, it’s essential that the retailer finds the right balance between goods and products that have traditionally sold well, new products for which demand is hard to estimate, and unique products that might provide an edge over competitors and bring people into their stores.

“For manufacturers the main challenge is being able to understand consumption on the store level. Modern services such as click-and-collect and next day home delivery mean retailers need to consider servicing demand from in-store inventory for e-commerce transactions as well, which has a knock-on effect for manufacturers at the opposite end of the supply chain.”

Simon Eagle, senior SCM consultant at SCALA believes there are many benefits to be had by adopting a demand driven approach. “Typical benefits from replacing a ‘forecast push’ process with the Demand Driven approach is achievement of planned service levels with up to 50 per cent less average inventory plus lower costs due to using less capacity and eliminating fire-fighting and expediting. In addition, far less focus upon forecasting is required, which frees up resource for more ‘value-add’ activities.

“Companies that are first to adopt demand driven in their sector are assured of a significant competitive advantage in terms of service, cash flow and cost.”

For Canitz, improving the management process for demand, no matter which model is adopted, brings its own benefits. “The upside to a more robust demand management program is far reaching. An improved forecast can drive greater collaboration and trust between partners, suppliers, retailers, and manufacturers. Companies are able to top line results while reducing the bottom line. Quantifiable benefits can be found in increased revenue of up to 20 per cent due to higher in-stock percentage of right products, and lower operating costs and improved working capital of ten per cent to 20 per cent due to lower inventory, less obsolete product, and decreased expediting. Combining higher revenue and decreasing costs drives dramatic improvements pre-tax Gross Margin of three to five per cent.

While it is easy to get carried away by promises of increased revenue and more streamlined operating costs, there are a few things to consider before implementing a demand driven model. Canitz cautions: “Traditional supply chains tend to have considerable latency related to long production times, lead times, inconsistent transport, and other inefficiencies. The concept of being demand driven is good, however every company has its unique operating challenges and strategies that drive many different supply chain configurations. No supply chain can be entirely demand driven and the ability to react in real-time in most supply chains is not required and could be very costly.”

Carter adds: “The increasing incidences of zero stock supply chain management puts pressure right through the supply chain, because companies have to make smaller more frequent production runs, fine tune their operations and work smarter. All manufacturers have the capacity to operate some zero inventory stock management, although it is clearly better suited to some market sectors than others. Ultimately, even the food industry can benefit, although this is operationally more difficult because of product lifespan and the need to manage best before dates.”


So, you’ve defined where a pull model may be more beneficial and you are looking to make changes to make your supply chain responsive and dynamic. What technologies and processes should retailers and manufacturers be considering? For Albert, it’s an issue of scalability. “Moving to the ‘pull model’ ensures that you are closer to the consumption point and better at sensing demand, and it will be important to ensure that systems are able to aggregate and compute the large amounts of data from within the business in real-time from across point of sale, e-commerce and even responses to marketing materials.

“As organisations become more confident with demand management systems additional external data sources may be considered, such as unstructured social media data, so it’s essential that systems are able to scale and provide the most complete overview of demand possible.”

Canitz calls for a holistic view: “To support higher levels of maturity retailers and manufacturers need to consider systems that enable management by exception, workflow and collaboration, what-if scenario analysis, and capabilities to manage products throughout their lifecycle.

“Systems should also enable quantitative and qualitative forecasting techniques to use intrinsic and extrinsic information. Finally, if you create a more accurate forecast but can’t use that forecast to improve revenue and reduce costs because of data integration and process collaboration issues the value we be lost. Therefore, companies should take a more holistic view of their supply chain planning solution requirements and only consider solution providers that provide an integrated suite of planning and optimisation solutions for demand, inventory, replenishment, supply, and manufacturing planning.”

Philip Ribbins, owner of supply chain software company Orchestr8, says: “ROI on supply chain technology has been very poor with little or no measurable business results, while technology has become ever more complex to meet the challenges of ever more complex supply chains.

“This, however, is a mistake – supply chain planning done well is not the result of clever maths or algorithms. Variability is unforecastable at a detailed level and always will be. Good plans cater for that and allow planners to operate in a simple environment to respond to the demand they are faced with on a day-to-day basis.

No right answer

“There is no one right answer – different planning and supply side set ups are required at different times, so having a solution that allows for that and supports decision making to select the right approach as well as catering for item level plans is crucial.

“Software needs to support the full business planning process – it makes no sense to conduct different parts of the planning process in different systems, and planning does encompass demand pull, forecast push, life cycle management, capacity planning and S&OP level modelling and decision support.

“The extended supply chain requires open and flexible platforms that can extend across the entire value chain regardless of the enterprise operating it. MRP-based planning will not support this environment because of the flexibility and extended nature of the planning process. Crucially, planning is best done outside of the execution systems.

“This gives the option to deliver real business value through multiple, appropriate supply side network planning, incorporating segmented supply side planning and cross enterprise set ups. In our experience, this transforms return on investment on system implementation from almost nothing (or even negative) to 500 per cent plus,” says Ribbins.

Carter recommends four essential actions for manufacturers to consider when looking to move to a demand driven model:

  1. Use a WMS to manage warehouse operations efficiently and estimate resourcing costs
  2. Change the business emphasis from making small numbers of bulk deliveries into distributors to a more customer focused B2C process, delivering a high volume of individualised orders
  3. Foster relationships with logistics providers and couriers e.g. two-man delivery companies who have the capability of delivering a high level of customer service to end consumers
  4. Expect to hold a level of just-in-case stocks if sourcing products from China with a seven – 12 week lead time and understand the minimum level of stock the business must hold to accommodate lead times and sudden peaks in demand – especially with cash cow products.
Share this story: