Continuous improvement

by Bill Gaw

The MRP evolution took us down the road of computer sophistication. It was supposed to be the panacea for solving all manufacturing problems. Little did we know that when we finally arrived at the final phase -- ERP -- that we would still be facing daily parts shortages, shop floor disasters and end-of-the-month chaos.

What happened to all those promises?

MRP/ERP are not complicated systems, at first look. We input a master schedule that uses bills of material and parts procurement lead times to calculate gross requirements. These requirements balance against the aggregate of on-hand inventory, work-in-process and open purchase orders to determine the net, time-phased requirements.

The resultant is subjected to lot size algorithms and planned orders are created. The final output is notification to planners in the form of action messages to either reschedule, reorder or cancel shop and/or purchase orders.

If we go deeper into what is happening in the gross to net requirement process, we find that many calculations are made based on the data and systems parameters supplied and maintained by planners. While a computer is flawless in its ability to calculate the answers, the data supplied by the planner is not. Consequently, the answers are subject to human error.

In our presentations, we do an exercise in statistical probability. Each participant writes down what he/she knows (or guesses) to be the percentage accuracy of their company's master schedule input data. To arrive at the aggregate input accuracy of the master schedule, they convert the percentages to decimal equivalents and multiply each to the other. Statistical probability is not the averaging of the decimals, as many people think.

Using the same statistical probability approach, the resultant decimal is used as the master schedule accuracy input into the requirements planning step to calculate a shop order launch accuracy. An accuracy level of 0.70 or 70 percent is quite common and indicates that order launching and rescheduling efforts are based on a system error of 30 percent. Is there any wonder why MRP and ERP are not the panacea we once thought them to be?

An effective method for evaluating how well a company is doing in managing their MRP input data is to ask questions as to how accurate are their bills of materials, how accurate is their purchase order status, how accurate are their inventory records, and so forth. If the answers  are vague -- good, OK, not too bad -- you know this is a company that needs help in stepping up to the problems of poor information integrity.

If a company is not measuring their system's data integrity, nor in constant pursuit of continuous improvement, its results will always be poor and its production environment will surely produce shop floor chaos and late deliveries to internal schedules and to customers.

For a measure of MRP/ERP shortcomings, one needs only to spend some time in a manufacturing facility, especially during the last weeks of the final financial quarter. In a typical company, you'll find that converting the quarterly financial forecast into reality still requires overtime, internal/external expediting, last minute on-the-run product changes and even a little smoke and mirrors. Results are scrap, rework and warranty costs that negatively impact a company's bottom line performance.

In addition, marginal quality and late shipments deliver less than acceptable customer satisfaction. Companies that spent thousands of dollars in pursuing MRP/ERP are devastated when they experience a business decline due to noncompetitive pricing caused by uncontrolled operating costs. Is there an alternative?

Certainly, we call it Kaizen-based lean manufacturing (KBLM).

Kaizen (pronounced ky'zen) is the Japanese word that means gradual, continuous improvement. In my experience, managing a continuous improvement project is difficult but a kaizen program presents a unique challenge. The kaizen program has no end. It is sustainable and successful only when management makes a commitment to stay the course. Discipline and tenacity are basic requisites for kaizen success.

KBLM is a proven methodology that employs practical tools and techniques that optimize manufacturing performance and help companies to consistently exceed performance expectations.

KBLM involves arranging and defining manufacturing resources so products flow most efficiently through the manufacturing process. Today, most manufacturing companies are still organized for functional manufacturing; mechanical assemblies, electronic boards, cables, machined components and purchased parts are produced or purchased in lot sizes and received, inspected and moved to stockrooms.

This process includes picking parts to fill shop orders and moving shop orders to the production-machining and assembly-build areas. When parts are completed, they are returned to the stockroom to be picked for the next higher assembly shop order. Finally the end product is picked, assembled, tested and accepted.

KBLM eliminates all non-value-added tasks in this order-launch-and-expedite system. The result is significant increase in quality, speed and profits.

No matter how much sophistication you add to computerized shop floor control systems, if you fail to master the eight basics of KBLM, you will never eliminate the chaos that grips your shop floor day-to-day activities.

Back to top

Back to Web-exclusive articles archives