Moving to Make-to-Demand: An Operations Overview

Executive Summary
To cope with increasingly volatile markets and shareholder expectations for continually improved financial performance, manufacturing companies are moving to Make-to-Demand (MTD). The goal of MTD is to keep activities synchronized with demand as it shifts. At the same time, it lowers operating costs to bring companies a competitive advantage. However, moving to MTD also brings challenges that can overwhelm the benefits if not managed well.

MTD requires a fundamental shift in strategies. Cornerstone concepts for MTD like building to order or to the previous day's demand; design for assembly; and supply chain visibility to changing conditions, designs, and costs are not as critical in a traditional mass production environment. As a result of these shifts in strategy, new business processes must be put in place. Lean, flow, and no-waste manufacturing principles are the foundation for these business processes.

Quality is more critical to on-time customer delivery when inventories are lean and products varied. As-built product history is often an important part of the customer shipment. In MTD, tracking and visibility become nearly impossible to achieve without computer-based applications. People simply cannot handle the quantity of data involved to make good decisions in Internet time. All of this points to the need for Manufacturing Execution Systems (MES).

Industry Directions research Becoming Demand Driven showed that plant-focused initiatives are the starting point and foundation for all other MTD activities. Those with the best financial performance in that survey recognize the value of plant floor visibility – not only to production but also to many groups in the company. Newer MES like CIMNET's Factelligence not only support effective MTD operations, they allow the manufacturer to implement incrementally, starting with their highest priority functions. This configurable approach also recognizes that each plant has different needs in moving to MTD, and rather than prescribing what may be overkill, allows the company and its consultants to decide what is needed.

Why are plant floor systems critical to MTD? For one, the visibility they deliver allows inventory reductions and improvements in resource usage. They also facilitate built-in quality at each step of the process, reducing waste and rework costs. An MES can also foster higher revenues – MTD companies often capture market share with their faster reaction to changing demand patterns, but would be hard pressed to have adequately flexible facilities without this type of plant-wide application. MES may not seem like the core of improving margins, but it does play a role in many ways.

The MTD World
Margins are tight and markets are uncertain today. As a result, manufacturers are finding it imperative to reduce inventories and move more quickly to react to shifts in demand. Some companies are moving to true make-to-order (MTO) models, where nothing is produced until there is a customer order. But even those making standard products are shifting to make-to-demand (MTD) models.

In MTD, current and changing demand patterns determine what the plant makes, while the long-term forecast mainly determines procurement contracts. The plan is finalized as close as possible to ship dates in most cases. The benefits include:

  • lower inventory carrying costs,
  • reduced risk of overstock or obsolete inventory,
  • better customer responsiveness and thus increased customer satisfaction,
  • increased "effective capacity" using the same resources,
  • greater flexibility to keep up with shifting demand and market trends,
  • higher gross margins and greater cash flow.
For companies that have been make-to-stock (MTS), moving to MTD changes everything. The basic strategies are different, which means that business processes and relationships must also change to achieve the goals. Information systems must support these new processes.

Strategies – MTD strategies aim to satisfy customer and market demand, no matter how fickle it is. They focus on coordination between departments and supply chain partners, responsiveness, speed, and flexibility. Examples of MTD strategies include: flexible product design or formula for customization; flexible facilities for production and assembly as well as logistics; deliver-on-demand; design for assembly; supply chain visibility to changing conditions, designs, and costs; and make-to-order (MTO) or build based on yesterday's consumption or demand. Many companies work with a consultant to prioritize the strategies that will best equip them to be leaders in their industry.

Processes – To execute these strategies, business processes must change. Design processes must be faster and more agile to meet changing demand. Production planning is shorter-term and based primarily on actual orders or consumption rather than a forecast set months in advance. Supplier relationships must shift to accommodate increased or decreased demand and the possibility of new requirements. Logistics is more critical, as the timelines are shorter. Customer care requires more order-specific status and as-shipped information. And production processes must change as a foundation for MTD operations.

Systems – Clearly, information systems must be more sophisticated to accommodate delivering this amount of information to the right people fast enough for them to react to demand shifts. Paper and spreadsheet systems are not adequate in the MTD environment.

In our benchmark 2000 study Becoming Demand Driven, Industry Directions learned that companies go through four stages in moving to MTD. They move from plant-focused to supply chain focused, then enterprise focused and, finally, customer-focused. Plant-focused is the foundation for all other steps. Best-in-class companies in our research were more likely than others to understand that visibility to plant operations is key to success with MTD.

We defined best-in-class companies as those farthest along the path to MTD and who consistently scored highest on metrics such as cost reduction, order fill accuracy, inventory turns improvement, and reduced working capital. The common factor among those who are best-in-class in the Demand Driven survey is a production issue: their cycle times are within the customer order lead times. In other words, these leaders can wait until they get a customer order in to begin production and still meet the desired due date. Their production and logistics operations are fast enough that they can make products to order without keeping customers waiting.

So key to MTD success is production speed and accuracy. The challenge is that this speed must be achieved in the face of increases in: complexity, inventory and order challenges, the impact of poor quality on customers, expectations for information as part of the product. Some details on this follow, which help illustrate that an MTD operation is too complex to manage without operational software support.

Operational software that delivers detailed yet plant-wide information is generally called a Manufacturing Execution System (MES). Newer MES offerings are modular – like Factelligence from CIMNET Inc. This configurable structure allows companies to pick and choose which functions to implement to best achieve their operating goals. The system can also grow to support changing strategies and business processes as they become more sophisticated.

Tracking Complexity
Companies moving to MTD may begin to track orders differently. In many mass production plants, work orders focus on meeting forecasted demand by product family. In MTO environments, every individual order counts, and they must be tracked within the aggregated production or work orders in the plant.

Many companies moving to MTD have used the principles of lean or flow manufacturing to streamline operations. Industry Directions' research on Flow Manufacturing states: "Flow manufacturing is a pull-driven strategy… At its core is the principle that demand can be synchronized to a daily production rate by properly sequencing items on a flow line that is replenished frequently by suppliers. In this way, inventory is kept to a minimum, goods are made to demand, cycle times fall within the required order-to-deliver response times, and constraints are minimized."

While some consultants originally encouraged companies to use only minimal software for flow, the current wisdom is that information systems can be useful for even a very streamlined operation. With today's expectations for order status on request, walking around to understand progress is not usually fast enough. Further, many customers want instant answers over the Internet, which requires the information to be on a computer from the outset.

And that is just the aspect of tracking orders. Tracking is even more complex at other levels:

  • Tracking materials is critical in many industries – whether for supplier quality purposes, product shelf life, inventory costing, or Vendor Managed Inventory (VMI). When the goal is to have low stock levels, the system must stay alert to materials problems.
  • Tracking resource usage by configuration is a key to understanding actual costs of configured-to-order products, or variants of standard products. Companies without this level of detailed cost data have been known to actively promote money-losing products.
  • Gauging processes & improvement trends is also trickier when the mix of products through a line increases. Detailed tracking allows visibility into the process effectiveness for each variant or SKU.
  • Gathering data on actual production for aggregate use in ERP and supply chain management systems can also be complex in a MTD environment. Actual-vs.-plan is the key to replanning, and in MTD, every product and variant – and possibly every customer order – may need to be tracked separately.
  • Recording full "as-built" and key quality records results in large amounts of data for each batch, lot, or individual unit produced.
Clearly, companies that simply let their lines run with basic assumptions of how much standard product will run through in a given time period will need to shift their mindset for MTD. Detailed data is the key to fine-tuning a lean or flow operation – and computer software is the best mechanism to accomplish that.

Visibility for Decision-Making
People will always be the key to sound decisions. However, humans have a couple of shortcomings in their natural style that indicate a need for information systems to support decisions in complex environments like MTD.

  1. Capacity – people can only handle so much, and then they become overloaded.
  2. Past experience – the tendency to rely on what worked before can help solve recurring problems quickly, but can also be misleading in the face of new situations. MTD means new situations are extremely common.
MTD increases both the number of relevant data points on which to make decisions and also the variety of conditions that may arise. Clearly, additional tools are required to help present information very clearly and only as needed.

In MTD environments, people need enhanced visibility. On one side, people need to understand the demand in more detail. Those working in a pure MTO environment need to see all of the actual orders. Others in a MTD pull operation need to see exactly what was consumed or ordered very recently – ideally the previous day or few days. In either case, the MTD operation breaks demand down into its constituent parts, rather than taking the aggregate view of a long-term forecast.

People in the operation also need excellent information to improve processes and understand where to invest. As variability in demand and thus variety in products increases, decisions become more complex. It is often less obvious in MTD where a new machine would pay off vs. making a change to the mix of products down a line. This type of analysis requires detailed data and experienced analysts.

Finance teams in many manufacturing companies are trying to build systems that will help them understand cost structures. These may be dashboard-style systems or performance management systems that work from a data warehouse or data mart. A good plant floor system can deliver the most timely and accurate data about the usage and costs associated with most of a manufacturer's assets – plant, equipment, personnel, and materials. So while finance may not walk around the plant, an MES tracking the details of the operation is the foundation for financial cost measurement systems.

Operations managers, from supervisors to plant managers to VPs, need a quick way to review the status of operations. New systems provide simple, often color-coded signals to show how things are going. Management can then allocate time and energy to fixing severe or chronic problems.

Examples of visibility include:

  • Production status including the exact location of each order with predictive indicators on order completion
  • Labor usage including the quantity and labor categories
  • Machine utilization including their Overall Equipment Effectiveness, status and downtime patterns
  • Quality and process problems by product, supplier, operator, and machine or production line
  • Materials disposition such as scrap, rework, materials review board (MRB)
  • Current CAD drawings, tooling instructions, and operator procedures.
The Quality Challenge
First-pass quality at each operation is ideal for manufacturing in any case, but even more critical in MTD when every glitch may impact customer order delivery. In a MTS environment, companies can simply pull another of the same item out of stock if one in process is defective. In MTD, the concept is to have only the stock you need to meet demand, with minimal safety stock. In a true MTO environment, each product may be configured specifically for a customer order, so replacement can be a challenge, particularly if the component materials are specific to a configuration.

Another reason why quality matters so much in a MTD environment comes back to the differentiator for best-in-class manufacturers: cycle time. Without high quality products and processes, it's nearly impossible to consistently achieve short production cycle times. Quality problems can significantly delay cycle times with materials disposition, rework, and re-testing times.

Incoming quality processes are important to MTD operations also. For ongoing improvements, the data needs to be in a form that the plant can easily send to procurement and share with suppliers. On a day-to-day basis, the incoming quality process can affect the overall cost of goods – and the ability to negotiate cost-effective contracts with the best suppliers based on their performance.

As mentioned above, quality & process trending is a tougher feat in MTD, since products are not identical. Detailed data needs to be analyzed to separate true quality issues from those caused by new variants and constant change on the line. The continuous improvement and process design elements of quality must be able to handle a higher mix and shorter cycle time environment.

Quality information about both products and processes is important as feedback to product design or R&D groups for improvements, also. Many companies are now putting renewed emphasis on product design to achieve market leadership. Since most of the cost of a product is determined at design time, plants with the detailed data about performance in production have the best chance of winning in the design game.

As-Built History
Manufacturers are increasingly finding that information is part of the product. Customers (and in some cases, regulators) want to know exactly what is in the product and how it was manufactured. In MTD, that often means creating a detailed product history for every product shipped.

For batch manufacturers, it's a batch history record and for discrete manufacturers, it's a genealogy of each individual product built. Usually in discrete, this ends up as a hierarchy of serial or lot numbers to indicate all of the materials included and their origin. Some business customers require detailed as-built information from manufacturers with the shipment of products. Even those that do not usually appreciate knowing the plant can produce this documentation.

The plant operation must capture genealogy information as it builds the product, or the process would become terribly inefficient. The plant mainly benefits from as-built history when it needs to rework a batch of products that used raw materials or assemblies with a specific defect. It's also key to compliance in regulated industries such as aerospace, pharmaceutical, or food. However, others in the organization will use the as-built history from a plant system on a regular basis.

One of the main benefits to the corporation of having this as-built history is reduced risk from liability exposure. In many industries, products have a warranty or the company makes extra revenue from servicing products. In these environments, the as-built history is key to delivering after-sales service. When one unit is fixed for a component problem, it may also suggest problems with similar units. Rather than recalling all of the units of a particular model, MTD companies with as-built histories can recall only those built from a particular batch of material. As-built detail history also allows manufacturers to share the burden of a recall or repair issue with suppliers appropriately.

Executives are always worried about maintaining profit margins also. Many MTD discrete companies have estimating departments that develop quotes for new orders. With the detailed product history, they can quote on new projects from a perspective of credible data on actual costs, time, and resources for previous similar products.

Making MTD Pay Off
Markets today are unpredictable, so MTD is actually a safer strategy than traditional mass production or MTS. When production is tied very tightly to actual demand, a company can react quickly to upturns and downturns. If the past few years have shown anything, it is that this ability to adjust a business to the market is key to financial success.

Industry Directions' research (Becoming Demand Driven) also shows that leaders in MTD perform better on key financial KPIs. They outperform others in achieving:

  • cost reductions,
  • inventory turns,
  • productivity improvements,
  • increased effective capacity without new investment, and
  • reduced working capital.
Moving into MTD requires new strategies, new business processes, and new information systems that can support complexity and variability. People's ability to make good decisions is quickly overwhelmed if they don't have reliable and detailed information on which to base them.

In MTD operations, this means a plant-wide system that goes into greater detail than ERP's production capabilities. This MES system should be able to track as-built configurations of individual products within a customer order as they move through the process. Internet web clients allow information access as today's employees and customers expect. The MES also helps ensure product quality, as it can assist in analyzing problems and trends, and pinpointing the true sources of problems.

In MTD environments, plant operations personnel need to embrace the system – so it needs to be easy to use. Further, the system must be capable of frequent change and reconfiguration to match changing business processes, product mix, and information needs across the enterprise and the supply chain.

Because each plant – even within a company – has its unique aspects, the software should be configurable. Further, since each company will identify different priorities or pain points for operations, the system should ideally allow customers to pick and choose which functions to implement and allow graceful add-ons

Fortunately, systems are available that support all of the MTD business process and foster the critical continuous improvement efforts of market leaders. Factelligence from CIMNET is a prime example of all of this. CIMNET is one of the first companies to combine expertise in both discrete and batch process industries, and to create an easy-to-implement product to serve that whole range of customers.

Companies cannot expect to transform mass production operations overnight, or use antiquated systems and sustain their process change. MTD requires setting appropriate goals and rewards, then creating business processes to match, and buying operations software that can support this level of complexity. Experience shows that the rewards are worth it.