The ability to reduce lead times is a key differentiator in competitive markets, service industries and the nonprofit arena. All Lean tools attack waste; and time, specifically lead time, is a key area where waste tends to creep in. Three Lean tools in particular can help: Value Stream Mapping (VSM), cellular/flow designs, and setup and changeover methods.
When we start thinking about reducing lead times, the first step is to build a VSM of the current state processes, recording what is really happening, not what is supposed to be happening. For example, at a printing company I worked with, the amount of time it took to process a new quote was excessive. There was a great deal of confusion, finger pointing and guesswork regarding the cycle and what should be done about it. So, my colleagues and I hoisted a video camera on one shoulder to find out exactly what was happening. We pretended we were a new quote request from a big customer.
By forcing ourselves to actually “be the quote,” we went through the various process steps, manual and electronic, and captured interviews and artifacts to record the current-state process. At each handoff, we generated a new data box and recorded the high, low and mean delay times for each, along with other details on cycle times and quality.
At the end of the process, our team analyzed the findings. It was shocking. From the time the phone rang to request a new label, to the time the quotation and camera-ready content was approved, we found 43 handoffs and 22 process delays, from minutes to days in duration. During this time, a little more than five hours of human effort was expended across an average of 38 working days.
By reworking the system, unnecessary delays and handoffs were eliminated. In a very short time, the team was even able to cut lead time by 50%.
The next step is to examine functional layouts to identify lead time reductions through better flow and cellular production. For instance, a paper-based insurance claims company wanted to examine how mail is handled from the time it’s received to the time the claims department assigns the tracking number and records it in the computer system for processing. The team first created a VSM and spaghetti diagram of the process.
The old process took more than four days and involved 10 handoffs and various delays. Each piece of mail received about 49 seconds of human effort on two different floors. Much of the delays and extra handling steps were necessitated by the functional layout with batch-and-queue piles of mail being pushed through.
In conjunction with scanners to replace micro filming, the team came up with a new work layout design that moved all the steps to one floor and established one-piece flow. Not only has the company seen a 20% productivity improvement, a given paper claim now moves through in well under two days.
Another example: At a point of purchase display manufacturing company, the decision was made to reorganize the value stream flow and layout. The old process involved mass-producing grids of welded wire mats in a large batch. They first were sent to a press brake department, then were moved to a mass-welding department, and at last were delivered to a final assembly area. The larger orders took about 8–10 weeks.
After analyzing the typical workflow, company decision makers installed quick disconnects for power, water and air in several areas of the plant. By eliminating functional layout constraints, the company can chain together several operations in a single flow line. This enables the larger orders—which make up 80% of sales—to be produced in a cell, instead of the old batch-and-queue approach. Today, the company commonly completes these large orders in just two to three weeks—a huge lead time reduction over the original. The beauty of this approach is not just in the collapsing of lead time; it also reduces scrap and increases productivity upward of 25%.
Plus, this is a business that experiences great competitive pressure from overseas. Reducing lead time facilitates more competitive pricing and faster order response, which is nearly impossible for overseas competitors.
Finally, we must attack a key source of long lead times. Long lead times often can be traced back to processing large batches first, before new orders can begin. After digging a little deeper, however, we often find the real root cause to be long setups or changeovers.
Quick changeover capability never seems to be given the attention it deserves at the design stage of a product or service. While spare parts, fittings, quick-change automation and redundant equipment are issues considered during development, they are too often disregarded during the actual development process.
Here’s an example. It costs an extra $100,000 for the appropriate fixtures, tools, automation, and equipment necessary to changeover a process in 10 minutes or less versus 4 hours. Let’s say the value-add of the process is $100 per hour to keep things simple. The smart accounting type will figure this out pretty quickly. There is zero payback for this added cost if we don’t need frequent changeovers. We know this by figuring the cost of a setup as $100 per hour multiplied by 4 hours per setup, which equals $400. Dividing $100,000 by $400 equals 250 setups. That’s a lot of setups required to make up the $100,000. Therefore, in the presence of uncertain market demands and product life cycles, it is very easy to justify cutting this corner.
Now, let’s say we offer 10 of these products or services and each is worth $10 million in annual sales, with 75% cost of goods sold (COGS) per product. This means about $100 million in annual sales with $75 million in COGS overall. They all share the same resource—say a large molding machine.
Now, with a four-hour setup, we aren’t going to do this every day for 10 products. That would be 40 hours a day in setups. We can’t even afford that in a whole month, as it eats up too much capacity. So, we probably need to run three-month batches of each product and try to cycle through them four times a year to keep up. This automatically forces the following results: two months inventory overall at a minimum, tying up more than $15 million in inventory at COGS, and a 12–16 week fixed lead time for changes in the schedule. This costs us about $3 million a year. See where this is going?
Okay, now what could have happened if we had spent the $1 million for the quick-change capability? What would be our payback if our only gain was reducing inventory batch sizes to two weeks each? Remember, we can now do setups for these 10 products in 10 minutes. This means we only need 100 minutes every two weeks, less than two hours of lost production time, which is even less than we needed for setups in the original scenario.
If we now have an average of one week of inventory, our average COGS inventory value is 1/50th of the $75 million annual COGS—$1.5 million in inventory, not $15 million. The differential in carrying costs is $3 million minus $300 thousand a year ($1.5 million times 20%). Hmm . . . that’s $2.7 million to cover the $1 million investment in quick changeover, a 2.7 to 1 payback in the first year. This doesn’t take into account any price or market share advantage the company will have in being able to quote and hold two-week initial delivery response windows versus up to 16-week windows.
The point of all this is, if we really want to compete on the basis of lead times, it’s important to make understanding and managing them a mission-critical activity in all parts of the business—especially executive planning and initial product and services development. Experts say that 75% to 90% of costs are fixed at the design stage. While you may not be able to re-engineer what you have today, there’s no excuse for continuing to repeat past mistakes tomorrow.