I Work in a Start up Company That Has a Very Complex Product Which Have Some Aspects That

Total Page:16

File Type:pdf, Size:1020Kb

I Work in a Start up Company That Has a Very Complex Product Which Have Some Aspects That

Table of Contents

Use of Lean Techniques in the “concept to launch” phase of development______2 Company Background______2 Lean Value Stream Mapping______3 What did we want to get out of the mapping process?______3 Drafts of the Value Stream Map______4 First Objective of the Value Stream Map______7 Second Objective of the Value Stream Map______7 Unexpected Results of the Mapping Process______7 Review of Current State Map______8 Future-State Map______8 Use of Lean Techniques in the “concept to launch” phase of development

There are many techniques developed to aid in the leaning of the process known as from “concept to launch”. To name a few:

Concurrent engineering Cross functional teams co-located in the same area Quality Functional Deployment (QFD) Design for Manufacturability/Assembly (DFM/A) Tollgate Design Review Process Design for Producibility Design for Experiments (DOE) Value Stream mapping

These are all interesting in themselves but I want to focus on Value Stream mapping specifically used in a startup company for a new process.

Value stream mapping is typically used in a mature product for the purpose of removing all of the non-value added waste. It can be used to set up pull systems, continuous flow manufacturing, re-layout manufacturing areas and many more.

Company Background

We have already established the cross-functional co-located team. Our team consists of 10 people, separated from the rest of the company in our own building. Our team consists three groups: R&D: Director of R&D, a senior scientist and an associate; Animal Studies group: Project manager and two people who work for her; Operations: Operations Director (me), an Electrical Engineering Manufacturing engineer, a manufacturing supervisor and 2 people working for her.

Our charter is simple. Develop a implantable medical device. The definition of success I as follows:

 Two back-to-back identical configurations built with a minimum of 1 month successfully implanted in rats.  Clinical trial Out of USA with successful results.  Clinical trail inside USA with successful results.  Commercialization.

Success is defined as 80% of the items work, 80% of the time. The work is divided up so that the R&D group develops the process until it is ready to hand over to the Operations group, once the process is repeatable, documented and standardized. The Operations group participates in the development as much as possible so that ideas can be shared.

The product is complex and is very well documented regarding the process instructions. Also, some of the processes use materials in unique and innovative ways and processes will have to be created to manufacture them.

How can the processes be designed so that they have a good continuous flow with few bottlenecks? How can we identify design issues early and keep track of them until they are solved?

Lean Value Stream Mapping

Most of the people reading this article will be very familiar with value stream mapping, so I won’t discuss the standard methods but only the additions or modifications made for this application and what we learned by going through the process.

In this example, only one of the sub processes will be explained and the benefits that were received by using this tool.

What did we want to get out of the mapping process?

During the manufacturing of the first product that was used for the back-to-back animal studies and the first clinical trial, a couple of things happened that highlighted the need to map the process.

With only 2 weeks left in the schedule before the product had to be shipped out of the country, we could not get satisfactory product out of the sieving process. This led to many late nights doing the menial tasks required to screen material, which was finally obtained.

Then in the last test before product was approved, it was discovered that the scientist forgot to perform the critical operation of poking holes into the material. This led to two rework cycles and of course many more late nights and weekends to finally finish the product.

From an Operations perspective, rework is the devil. As soon as rework starts, one can be sure that errors will come along with it. And of course, it did. The yields that had been achieved in the back-to-back studies of 86% for the first test and 83% for the second, had dropped to 71% and 32% respectively. And we didn’t even know what was causing the problems. Luckily we were able to make the required quantity for the test and get them off to their destination in the required time.

What did we want do get out of the Lean Value Stream Mapping process?

Number One: The first thing that we wanted to get out of the Value Stream Map was to see an accurate picture of how long the process took taking into account the expected yields.

Number Two: Put into place checks and balances so that if one part of the process were not performed correctly, it would be caught at the next step.

Drafts of the Value Stream Map

The format of the map would determine what information was collected and therefore what answers were obtained. The purpose of the value stream map at this point was to take a picture of one lot of material as it went through the process. So the stages between each process would calculate the output from the previous process, not track the amount of inventory that had collected at that location.

The following categories were added to each information box:

Categories added to the Process box:

 Location of the process:  Design Issues:  Performed by:

Categories added to the Tracking Box:

 Cycle time:  Expected Yield:  Actual Yield:  Critical Process: Yes or No  Shelf Life:

Critical Process is defined as a process where a test or inspection is performed which would indicate there is a problem causing the operations to be stopped or delayed.

There is space at the bottom to add any for suggested process improvements and the decisions that were made for the Future State Map. Also, notes can be added that provide extra details regarding how process time is broken down, etc.

See Figure 1 for an example of format for the map. Title of Process Title of Process

Location - Location - Design issues - Design issues -

Input quantity Output Quantity from supplier Performed by - Performed by - Output Quantity

C/T = C/T =

Expected Yield = Expected Yield =

Actual Yield = Actual Yield =

Critical process = Critical process =

Shelf life = Shelf life =

Value Added = Add Time to Complete Process Time to Complete Process “time to complete” at each process

Suggested Process Improvements: Suggested Process Improvements:

Future State Map: Future State Map:

Figure 1

The map was constructed using the manufacturing process instructions and gathering information. The next step was to get with the team of the scientist, manufacturing supervisor and myself to review the map, verify the information and add anything that had been left out.

During the meeting, the supervisor asked about the cleaning process and why that wasn’t in the instructions. The Director of R&D joined in with more information about the lead- time and the fact that he is experiencing ESD problems when he is screening. The supervisor discovered that the operator wasn’t following the instructions. It turned out to be a forum for discussing the design as well as a road map to record the process as it is being developed.

See Figure 2 for an example of the map as it was filled in, it only shows two processes but the rest of them are listed below: Baking Sieving

Location - Sieve room Location - Sieve room. Design issues - Process is Design issues - None too variable. 1 lot = 1.5 box 15 boxes (5 lbs/9 (250 grams of 212 cups) Performed by - Mfg. Performed by - Mfg. 2 trays = 2 boxes microns)

C/T = 6 cups .5 hr; 1 C/T = 4-48 hours box = 5 hrs. EY = 100% EY = 80%

AY = AY =

CP = Bi CP = Yes

Shelf life = After open Shelf life = within 1 .5 hr mo. In dessicator, 2 days if not

4-48 hours Value Added = 1 day 8 hours (1 day) usually overnight and 1 night

Special handling: end of day .5 hour for cleaning. Test after 2 cycles - go/no go

Suggested Process Improvements: Put 6 Suggested Process Improvements: None cups worth into one container to eliminate handling of hot materials. Future State Map: Purchase a new design of siever that will eliminate the test after 2 Future State Map: cycles, improve reliability and decrease time.

Figure 2

The key processes are:

1. Baking 2. Sieving 3. Substrate 4. Visual Inspection 5. Testing 6. Preparation for screening 7. Preparation of solution, which flows into Screening onto Substrate

First Objective of the Value Stream Map

The first objective is to see an accurate picture of how long the process took taking into account the expected yields. So, to get a picture of the whole process flow, the actual times are added together to come up with the process time.  It takes 1.5 boxes of raw material to yield enough material to start the process for making sheets of foam, and takes one day  The next step yields 12 sheets of foam and takes one day  The next step yields 8 out of the 12 sheets pass the inspection criteria and takes 1 hour.  The next screening process yields 4 out of 8 sheets and takes 1.5 days.  At this point the material can be used for the final process, which yields 4 out of 4 sheets and takes 1 day, but it cannot be done until at least 16 sheets are ready.

So it takes 9 working days to get enough material to go to the next process.

Second Objective of the Value Stream Map

The second objective is: Put into place checks and balances so that if one part of the process were not performed correctly, it would be caught at the next step.

By having a visual representation of the process, it was easy to put in place checks and balances for each operation. After each process either a test or a visual inspection was added to make sure that the previous process was performed correctly.

Unexpected Results of the Mapping Process

By having a visual representation of the process flow, it indicated that a pull system or a rudimentary supermarket could be put into place so that enough material would be available when needed.

At the beginning I was not expecting to be able to put in a pull system so early in the development phase, but it seemed like a perfect solution to our problem of running out of material and being able to highlight problems when they occurred since they would not be meeting their expected output.

Then we took it one step further based on Theory of Constraints and identified the constraint and made the buffer stock located before the bottleneck. The bottleneck is between process 4 and 5. A double bin kanban system will be set up at that location. When the first bin is empty, it will signal the production team to make more. We decided to use 16 discs required in each kanban.

Also, it seemed obvious that it was necessary to always have a snapshot of the actual yield, not just relying on the gut feel of the expected yields that were acquired from the Chief scientist. So, a form was created to follow the material through the process so that for each lot, the actual “passed” material would be recorded and we would know what the actual yields were and how long it took to make enough material that could be used in the next process. Review of Current State Map

The penciled current value stream map was laid out on the table when the Production supervisor, senior scientist and the manufacturing engineer entered the room for the review. The scientist said, “Wow, this is great we are using tools for looking at production instead of gut feel”.

There was a lively discussion about each process step. We checked the numbers that were used for yield and output and what the output was at the end. Then we discussed what would be required in order to make 80 units (our current largest production run was 10). We calculated that it would take 20 days to make 80 units.

Future-State Map

Now that we have a visual representation of the flow, we can see which operation takes the most time and start to work on ways to improve. Can we buy equipment, which is faster? The future state map is created during the creation of the current state map and added at the bottom of the chart, as shown in Figure 2.

Another thing that can be looked at easily now that we have a value stream map, is what design problems we have identified. The map can be used to track the problems, to get status on their resolution and any new problems that might arise.

Conclusion

It was a very satisfying experience for all. The team was so excited by the fact that Operations was getting deeply involved with the process while the product was still in development, that they expressed fear that they might lose the resources for this process to the other department in the company. Everyone expressed an interest in continuing with the rest of the processes that exist in the product.

Recommended publications