How to Achieve Lean Manufacturing
Written By: Predrag Jakovljevic
Published On: February 15 2006
How Should One Go About Achieving Lean?
Part One of this series provided one definition of lean manufacturing (see Lean Manufacturing: A Primer). However, another way to encompass the lean philosophy is Womack and Jones' 1996 definitions of the five main steps to achieving lean transition for enterprises. These steps are now widely accepted as a sort of lean manufacturing life cycle. The following are the five main elements of enabling a lean approach.
- Define value
- Map the value stream
- Make the activities flow
- Respond to customer demand
- Continuously improve
This is Part Three of a multi-part note.
Since it is lean manufacturing's role to deliver value to the customer, the first step for any manufacturer that attempts to make its organization lean is to define value from the perspective of the customer, whether the end customer or an intermediate customer. This value must be identified and expressed in terms of how the specific product meets the customer's need, at a specific price and at a specific time. To do that, one has to be able to evaluate performance in terms of customers, products, profitability analyses, and so on, by measuring well thought-out key performance indicators (KPIs), such as customer sales, product sales, profitability by customer, profitability by product, etc.
Map the Value Stream
As the next step, manufacturers must identify and map those activities that contribute to value and those which do not. The entire sequence of the activities or processes that are involved in creating, producing, and delivering a good or service to the market—from design and sourcing to production and shipment—is called the value stream. For any finished good, the value stream encompasses the raw material supplier, the manufacture and assembly of the good, and the distribution network. For a service, on the other hand, the value stream consists of suppliers, support personnel and technology, the service producer, and the distribution channel. The value stream may be controlled by a single business or by a network of several businesses.
Once the activities have been identified, companies must determine what activities are value-adding, what activities are non-value-adding but essential to the business (e.g., payroll), and what activities are non-value adding and non-essential to the business. The impact that necessary, non-value adding activities have on the value stream must be minimized, while non-value adding, non-necessary activities must be eliminated from the process. To that end, a value stream mapping (VSM), which is a logical diagram of every step involved in the material and information flows from the order to the delivery of a product, can be done for the current process and the future process. A visual representation of every step in a process is thereby drawn and key data, including customer demand rate, quality, and machine reliability, are noted down.
Because VSM can lead to potential cost reductions, improved throughput, higher asset utilization, etc., some software vendors are now providing business process design, viewing and publishing application tools, and even business process reference models or templates of best practice process models, to assist with VSM creation. These tools are used to design processes, to communicate them, and to educate and work according to decided processes. By modeling the processes, one can visualize what the organization does and how it does it, as well as gain a view of responsibilities. Meanwhile, by connecting tools and documentation to processes, one can visualize which activities are performed and who or what controls them. Solutions might also include deployed functionality for publishing processes, connected applications, and documentation to an intranet-based workplace, which would ease the communication of changes throughout an organization and support employees working according to decided processes. Processes could even be published to a Web site, which would make it far easier for employees to easily access suggested or decided processes.
Make the Activities Flow
Once manufacturers identify value-adding and non-value-adding but necessary activities, they should then work to make these activities flow as an uninterrupted movement of products or services through the value chain to the customer. This requires manufacturers to eliminate functional barriers and to develop a product-focused organization. Dramatically reducing lead and cycle times, and eliminating work in queue, batch processing, waiting, scrap, and unnecessary transportation in this way, should lead, in the best-case scenario, to single piece flow.
Several tools can help organizations to achieve flow manufacturing, including total productive maintenance (TPM), leveled schedules (heijunka), long-term purchase agreements with suppliers, just-in-time (JIT) call-offs (possibly via electronic kanbans), and supplier managed inventory (SMI) for raw materials. In addition, though it may be surprising to some, forecasting can also be important in lean manufacturing, as it provides the basis for long-term purchase agreements with suppliers and also determines long-term capacity requirements. Forecasting also helps with generating a leveled master production schedule (MPS) based on the flow of materials through the supply chain or factory and costs. This is particularly useful where there is variable demand or new product introductions (NPI), since there may always be one major capacity constraint or minor ones may pop up here and there.
In fact, while forecasting might have had a poor reputation in manufacturing circles (particularly among those firms attempting lean), recently there has been an increased awareness that with good collaborative planning and forecasting software that supports collaborative sales and operations planning (S&OP) processes, many manufacturers could improve their business performance (see Sales and Operations Planning). Thus, as with production planning, manufacturers need to remain on top of forecasting by leveraging much shorter review intervals than the traditional quarterly (if not yearly) updates. By taking forecasting more seriously and supporting it with smart, interactive tools, all parties within the manufacturing business should be on the same page at the end of the day, which should result in increased agility. The exception, of course, is manufacturers in volatile markets or with products with short lifecycles, for whom forecasting based on history often means missing the true demand signals from customers or distribution channels.
Flow manufacturing does not address synchronizing around the supply chain, multiple partners, and suppliers, since it is merely a shop-floor execution tool. If only for this reason, enterprises should still use supply chain planning (SCP) for strategic purposes in which multiple departments (sales and operations, inventory, distribution, collaborative demand management, transportation planning, etc.) are involved, such as planning for resources across an organization, preparing for promotions, negotiating long-term contracts, establishing objectives, and coordinating multi-site operations.
Respond to Customer Demand
After manufacturers remove waste and establish a seamless flow, they must transform into demand-driven organizations where customer demand pulls products through the value stream, driving manufacturing activity and material flow. The ultimate goal is to become so responsive that products are delivered only when the customer, internal or external, needs it (i.e., places or signals the actual order)—not before and not after, though delaying the use of material and labor as long as possible.
To that end, intuitive and visible pull signals should initiate manufacturing and movement of material. This might necessitate support for various types of pull signals or kanbans (e.g., printed-out cards or electronic signals in case of using computers). Production of product can consequently be order-less, with backflushing to report automatic material consumption of components. This should result in major reductions in work-in-process (WIP) and production cycle times. The number of cards to print and use for each item and receiving location can either be entered manually or calculated and updated automatically in an enterprise system.
Kanban cards contain somewhat differing information for production cards, which signal that the container should be filled with new parts (i.e., item number, item description, requesting location, packaging type, container quantity, and replenishment lead-time) and transportation cards, which signal that the container should be moved whether empty or full (i.e., item number, item description, sending location, receiving location, total number of kanban cards, and standard container quantity). In the so-called one-card type of kanban system, only production cards or transportation cards are used, while in the two-card type of system both production cards and transportation cards are used.
Taking it a step further, one could leverage product technology that fosters collaborative agreements with suppliers and trading partners to have signals and the requisite data collection as the actual products flow through the supply chain. This would increase transparency and avoid stock bullwhip effects across the entire supply chain. A good example would be a repetitive scheduling technique like customer delivery schedule (CDS), which supports repetitive demands for one or several items to one or several locations in one transaction for discrete times or time buckets.
Such schedule level management tools can enable users to receive demands with different levels of accuracy or validity from different partners within the customer organization, while managing different types of demands, such as consolidated customer forecasts, customer forecasts, customer call-offs and JIT call-offs, customer sales reports, and sales statistics. The tool can be either integrated with various electronic data interchange (EDI) components that support several types of EDI message standards and message types, or managed manually. Use of vendor managed inventory (VMI) or point-of-sale (POS) data on the customer side should have a similar effect.
Software vendors also offer a number of other tools to help organizations respond to customer demand. For instance, a supermarket is a tightly managed amount of inventory within the value stream that allows for a pull system. Such inventory buffers can contain either finished items or WIP. They are used to handle finished goods inventories that are replenished by a continuous flow pacemaker process, which falls somewhere between a continuous flow process and other manufacturing processes shared by other value streams, as well as for incoming parts and material being pulled from supplier locations.
Some vendors offer strong JIT call-off management functionality as a planning and execution environment for both proactive and reactive sequence deliveries, as well as for frequent electronic kanban deliveries. Using this functionality, final products are broken down into specific items, a specific end product is pinpointed and identified via a sequence number or a production identification (ID), and the sequence call-off is sent out when the final production sequence is frozen at the customer line (which may be either hours or days before the parts are needed, depending on the application).
For advanced flow-oriented planning and execution environments, support for supply in line sequence (SILS) is available. SILS promotes the use of configuration within sequence flows, which allows, for instance, the ordering of a configuration of cables for a single vehicle on the customer's production line. To be able to pack according to all the different pack demands that can be found within the SILS concept, a very flexible set of rules must support manual or automatic packing of call-off demands according to a variant of the SILS concept or the customer's demands.
However, the really fundamental process in sequencing is the capability to translate the real demand originating from the original equipment manufacturer (OEM) assembly (in the form of a broadcast EDI signal) into the configured item(s) required, along with the delivery sequence and the location where they are to be shipped. To that end, JIT sequencing software tools provide a direct link with the assembly line, taking into account last minute changes to the production schedule, while still operating within acceptable delivery parameters.
This visibility allows comparison of OEM requirements with the execution plans, and their adjustment as necessary. Any changes in requirements are rapidly identified, planned for, and communicated to production and down through to the lower tiers of the supply chain. This has the immediate effect of minimizing the amount of inventory buffer stock. The speed of the overall process provides the additional time needed to react and take effective action. The bottom-line benefit should be shortened lead times and reduced buffer stocks, which in turn reduce capital inventory investments with their attendant carrying costs and obsolete material write-offs.
The lean transformation is never truly complete. As the lean literature often says, lean is a journey rather than a single event with a final destination. Thus, the fifth step to achieving lean is continuous improvement in pursuit of perfection. Manufacturers, on an ongoing basis, should re-evaluate the value they deliver to their customers, as well as evaluate their entire value stream in a continual effort to improve processes and reduce effort, time, space, costs, and mistakes. The objective is to relentlessly continue to remove non-value adding activities, improve flow, and better satisfy customer demands across the entire supply chain (see Technology Enablers for the Lean Supply Chain).
To do that, the enterprise, as in the first, value-defining step, must evaluate performance by measuring KPIs. Key components here are built-in analytics that track metrics monitoring the pertinent performance. For instance, Intentia has its Opportunity Analyzer (OA) solution, which uses a top-down map, target, and plan approach to help with opportunity ranking, prioritization of alternative projects, and calculation of potential return on investment (ROI) and payback time. Business process improvements are supported by focusing improvement projects, measured by relevant KPIs on the lowest hanging fruits, and then implementing best practices so as to secure strategic management objectives (see Enterprise Process Improvement Software: Customer and Software Vendor Collaboration).
In addition to supporting the above technical lean tools, one should never forget the importance of achieving core lean principles via well publicised cultural and managerial initiatives, such as employee involvement, quality at the source, team responsibility, a flexible workforce, employment stability, system and process thinking, procedural discipline, open communications, continuous improvement, and continuous learning. In this regard, it is important to reiterate that lean is a transformational exercise that requires an organization to cast aside long-held beliefs and business processes, especially those related to the role of management and of workers. For instance, the traditional idea of keeping all available resources busy all the time remains appealing to executives, because of the seductive fallacy that it is possible to make a great deal of money just by keeping everything and everyone running, no matter what.
Because the cultural and managerial impact of a lean transformation is significant, any discussion of these supporting elements is outside the scope of this series of articles and is better left to organizational specialist likes of the Lean Enterprise Institute (LEI) or John Costanza Institute of Technology (JCIT). In any case, in the ideal manufacturing environment, process improvement and the skill level of the workforce would lead an organization toward an optimal point of production efficiency. In such cases, the following five zero principles best describe the practical application of Ohno's vision (in the order they would most like be achieved): zero setup or changeover, zero defects, zero lead time, zero queue (and buffers), and zero inventory.
This concludes Part Three of a multi-part note.