Customers Drive Desires Home:
How to Synchronize When Mass Customization Comes to Roost
By Julie Fraser
Henry Ford would be an utter failure telling today's customers that they can have his cars in any color they want, as long as it's black! In every sector, customers demand a huge variety of products now. To take just one example, you could get a different style of Walkman-like personal headset stereo for every person on a Christmas shopping list and no two would be the same: pink for 10-year old Diane; fluorescent green, unbreakable for 13-year old Paul; one with a CD player for 18-year old Chris; radio-only for Mom, etc. To customers, this looks great: one idea with variations for every need.
All of this variety looks more ominous if you run the manufacturing plant. The operations in the plant now making not just one, but 15 styles of personal mobile stereo have become exponentially more complex. At the same time, it drives down product lifecycles and cycle times dramatically; one major electronics manufacturer has a stated goal of shrinking cycle times by a factor of 10 within the next five years.
Alvin Toffler calls this "demassification" and points out that the complexity of the information overwhelms the advantages of economies of scale characterizing our business decisions for many years. Henry Ford's mass production plant could never accommodate mass customization, and many plants are still working from economies of scale.
Few manufacturers are immune to this change. Specialization of products to suit customers' tastes is happening not only in consumer products, but in all types of manufactured product industries. A supplier is no longer judged simply by cost and quality, as we are all looking for suppliers that can add value or customize the offering in a way that provides us with a competitive advantage. The result: more product variations, each with smaller production runs and shorter lifecycles. This proliferation of products can wreak havoc in an operation that's unprepared, which nearly every company is in one way or another.
Business Impacts
Let's look at the areas in which the proliferation of end items complicates a manufacturing operation:
Demand Swings: With more variants, it becomes very difficult to predict exactly what mix of orders will come in. Even if the volume stays reasonably level, it's tough to forecast production or material requirements due to the many possible end-items in a mix.
Data Overload: Describing the exact specifications of each model, style, or variant creates huge volumes of detailed data coming in and out of order entry, product engineering, materials and resource planning, and operations routings and work instructions.
Capacity Killers: With so many different styles, lot or batch sizes going through production, operations shrink, meaning more setups, changeovers, washdowns, and complexity to manage.
Inventory Issues: There are more inventory items to manage in most cases, because each variant may require different raw materials. To reduce end-item obsolescence, many plants build or produce up to a sub-product level, which creates semi-finished WIP to track and optimize on-hand levels.
Resource Competition: More end items are competing for the same production resources (machines, tools, people and materials). Thus, scheduling and sequencing become much more complex in order to optimize order fulfillment times while minimizing non-value-added setups between lots or batches.
In fact, mass customization sets up a real conflict: Customer service vs. profits. Each operation must balance between providing specific product configurations that add value for each customer and keeping costs low enough that the specialized products are all profitable, even in low volumes. Generally, non-value-added activities rise as product mix goes up: procurement costs for more raw materials; engineering time to create product specifications, routings and operating procedures for each variant; changeovers happen more frequently; and the overhead of tracking all of that data can be enormous.
System Supports
Fortunately, leading manufacturers and their system providers are working together to meet these challenges. The good news is that computing technology's geometric price performance increases are permeating from the control instrumentation up to operators, and throughout the planning offices. Even small plants can now afford the computers and controls to be competitive and provide customized products.
Demand-management: A few products are aimed at improving visibility into future demand. Current studies are showing that forecasting should really be done only at the point of purchase; with that end-demand forecast, distribution centers and plants can better calculate and plan. For this to work, the end customer must be willing to create a good system.
Data Handling: Here's where basic computing technology is paying off. Some software can simply take advantage of increased hardware capabilities. Products such as Product Data Management, Product Configurators, and SP-88 recipe management in batch industries help manage the complexity of product specifications and engineering changes. While simulation models took months to build and were often never maintainable 10 years ago, today we can build complex models of operations rapidly and keep them current with operations changes.
Capacity Scheduling: While capacity planning and finite capacity scheduling tools have been around for years, they often only consider capacity in the aggregate. This adds to infinite capacity accuracy, but to maximize capacity, schedulers must consider setups, sequencing, and other time-phasing issues. Through human innovation and capital investment, manufacturers have learned how to perform setups orders of magnitude more quickly than ten years ago. However, even with those improvements, the huge mix of products means most companies need sophisticated software to perform order sequencing, grouping and batching to minimize the frequency of changeovers while ensuring orders are delivered quickly.
Materials Time-Phasing: Plants have been working for years on replenishment and supplier strategies to raise their inventory turns and reduce stock levels. Further huge inventory reductions come when materials and resources are planned simultaneously. This is the only way to synchronize the operation: materials ordering is time-phased with resource scheduling and coordination. The output of this joint materials and resource schedule is a job sequence that balances capacity killers against customer order delivery times.
Optimized Resource Deployment: The most critical, and perhaps rarest toolset right now, is one that can ensure everyone and everything is working on the right order at the right time to speed the cycle time and increase value-added time. As pointed out above, a good schedule must not only consider capacity, but also detailed availability of resources and materials over time. This means having status input from plant operations, plus a detailed model of equipment changeovers and downtime, delivery timing of materials, and operator shift, break and holiday calendars. It must also consider all of the interdependencies between resources: it may only be possible to run a new style on one of five machines with a single set of tools, and operators specially certified to work with the specific raw material or machine/tool combination. Plants with this type of software can minimize inventories, maximize effective capacities, and reduce production cycle times. The result: higher return on assets.
Adding a set of innovative manufacturing software to current methods and systems can improve a company's chances of being truly "customer-driven" in this era of mass customization. The tools span from long-range demand planning to detailed resource synchronization, and from engineering product configuration to customer shipment and available-to-promise quoting. At the core is the ability to synchronize manufacturing activities to the market demand, taking into account that the demand mix, materials levels, and our internal resources all change from day to day.
How Close Can We Come?
A few examples illustrate the changes manufacturers are making to accommodate the need for end-product variety. Let's begin with a large manufacturer of knock-down furniture. This company used to be in the make-to-stock business, but with the help of manufacturing synchronization software, has now managed to switch to assemble-to-order production. Lot sizes are eight times smaller than they were in 1987, and each lot may translate into as many as 80 work orders on the shop floor. Average run times at a work center dropped from 6 hours to 50 minutes, significantly increasing the number of setups needed. Because of its ability to handle this complexity, the company's total sales have increased by over 60% in two years.
Another example is an Alcan plant that makes building products such as aluminum siding. The plant's product diversity has increased by fourfold. Again using manufacturing synchronization, lead time has shrunk from 28 days to five days. Total inventory has dropped by 40%, freeing floor space and dollars to invest in new machinery.
As these examples show, software exists today to handle many of the complexities that mass customization presents. While software vendors handle a multitude of functions, most now admit that they cannot do everything; alliances with specialized solution providers are key. Manufacturing synchronization and smart scheduling software play a key role in optimum plant resources deployment; configurators and product data management handle variety for discrete plants. True object-oriented software will provide more useful integration, as objects that reflect a real entity can be shared by many applications.
Even with the technology, manufacturers must be streamlining their procedures and thought processes. Manufacturing synchronization, or bringing all of the resources and materials together at the appropriate moment to deliver a customer order with special features right on time - neither early nor late - is a different discipline from most plants' current planning process. As the mix becomes less and less certain, manufacturers will come to depend on the day-to-day execution systems to ensure their profitability. Traditional planning systems are not intended to respond to the speed of change and responsiveness that mass customization demands.
Manufacturing is in a transitional period, learning to move from transactional systems to OLTP and decision-support. Further, we are learning to graduate from simple visibility of status and problems to being able to overcome a problem intelligently and rapidly. To remain profitable, we must handle complexity efficiently, but still use current investments in equipment, people and materials suppliers. It's still a difficult balancing act, but companies facing the need to move into mass customization are finding useful solutions.
Julie Fraser is senior strategic advisor for Berclain (Schaumburg, Ill.), a provider of software and methods for manufacturing synchronization & scheduling to discrete and batch process manufacturers. She can be reached at (508) 362-3480, or by e-mail: [email protected]. Reproduction by any means prohibited.