Europe is currently experiencing a refugee crisis. Hard numbers are difficult to obtain, but it is estimated that one million refugees arrived in Germany in 2015. Government authorities were ill-prepared to handle and organize these people. Significant resources have been put in, but they never seem to be enough. The organizational processes are not yet functioning well.
As it happens, I am an expert in improving organizational and other processes. I decided to help, and together with two other professors, founded an initiative, Lean for Refugees. We are politically neutral, volunteering our time to organize these processes so we can help both the refugees and the government. Let me give you an overview of what we have done so far.
Who We Are
We are three professors and a few volunteers who spend our time helping the authorities organize the processes around refugees. For this we started a small initiative, Lean for Refugees (website in German).
|Prof. Dr. Constantin May|
Ansbach University of Applied Sciences
|Prof. Dr. Christoph Roser|
Karlsruhe University of Applied Sciences
|Prof. Dr. Andreas Syska|
Niederrhein University of Applied Sciences; www.faszination-produktion.de
What We Are Not
The refugee crisis is dividing Germany and Europe. Some people would like to let everybody in, and then some more, whereas others would prefer to build a wall that puts the Berlin Wall to shame. Society, with the help of politics, has to come to a consensus for this problem. However, this consensus is not a problem we can help with. We are experts on lean, not politics.
So, first, we stay neutral, and, more importantly, stay out of the discussion of who can come in and who not. So, please no hate mail for being too left or too right! In any case, regardless of what the consensus and the laws are or will be, they have to be implemented. And that’s where we come in.
Second, there are limits to what we can do. As you will see below, some laws are difficult to implement. Different software systems are not always compatible, bug free, or user friendly. We cannot change laws to suit the process. We cannot fix or change a Germany-wide software program. But even with these limitations, there is still lots of potential for improvement.
Finally, we do not aim for perfection. Our goal is merely to make the situation better than it is now. We would rather take many small steps in the right direction than one big step for a supposedly perfect solution that does not work in the end.
Who We Worked With
Altogether we spent almost three days in the LEA Mannheim (Landeserstaufnahmeeinrichtung, State Office for initial registration). This is one of the state of Baden-Württemberg offices that gives refugees their initial registration papers called BüMa (Bescheinigung über die Meldung als Asylsuchender, Confirmation of Registration as Asylum claimant). These are sort of a ID substitute for refugees. With those, the refugees are entitled to benefits and they can apply for asylum, which is a separate process.
There are approximately 250 to 500 refugees arriving in Mannheim per week, which is also approximately the speed at which the office can hand out papers. However, since the capacity was ramped up just recently, there is approximately a 5-week backlog of unregistered refugees in Mannheim (and similar in most of Germany), hence the need to improve capacity. The actual asylum claim process has a much larger backlog of multiple months.
The Current State
We started (as with pretty much any lean activity) with understanding the processes. The refugees are transported by bus from their lodging to the LEA and then back. The actual process has three major steps:
The first step selects groups or individuals from the mass of refugees waiting. The registration is done, if possible, in family groups (father/mother/uncle/aunt/children…), but also individuals without relatives. The challenge is to determine the relation between the people, their ages, and if they have already been registered. A photo is taken and the number of people is added in a database, “EASY,” that aims to distribute the refugees evenly across Germany.
In a second step, the actual details of the refugees are registered. Name, origin, language, date of birth, religion, etc. are first written down using a multilingual registration sheet. Afterward this info is put in a software system, MigVIS. Unfortunately, this is not standardized across Germany; different states and federal agencies have individual and incompatible solutions.
With the ultimate accuracy of the German bureaucracy, this results in 17 pages of print outs that have 24 seals on them, each of which has to be signed by the refugee. Luckily, for every additional refugee in the group there are only 4 more pages and signatures per person. These are the registration papers (BüMa), a consent form in German, a consent form in their native language, and bilingual legal instructions, all of which have multiple copies for the refugee, the federal office for refugees, the state office for refugees, the health office, the state government, etc.
Photo & Fingerprints (Erkennungsdienst)
The last step before the refugee gets his papers is another photo and fingerprints. The photo has to be without head scarf, which is a problem for some women. The fingerprints are easy, as most refugees have lots of practice in giving fingerprints. According to German news, some refugees get fingerprinted up to 8 times by different agencies while in Germany.
Current State Summary
Based on what we observed, and together with the employees from the LEA, we created a swim lane diagram, including problems and potentials. Like most value stream diagrams, these are only useful for the people who made them; for others, it is only to impress. Hence, be impressed! On a side note, I usually prefer to have no more than 15 to 30 blocks on a value stream map to keep it manageable. Yet, this was moderated by another volunteer, and I do not mess with other people’s style of moderation while doing a workshop. In any case, the results were good (and impressive 🙂 ).
We found lots of improvement potentials (even without changing software or laws 🙂 ).
- Many processes had a lot of waiting time. Especially the photo and fingerprint was 30 minutes of work, followed by 30 minutes of waiting for the next refugees.
- The refugees were brought into the offices from the waiting rooms 4 or 5 times. Each time involved a lot of searching.
- There were no visualizations for the refugees, to explain the process and guide them along.
- Many of the refugees arriving by bus did not meet the conditions for registration (i.e., accompanied minors need a statement by the youth welfare office, all family members that are fleeing together have to be present, …).
- The whole process is a batch process for multiple groups of refugees.
The Future State
Based on this, we developed a future state. Using a conservative estimate, we believe that we can improve throughput by at least 50%. The most important improvement ideas are below:
- Most significantly, we split the processes in smaller steps that are easier to learn and provide a better-balanced workload.
- We divided the steps into front-office and back-office steps. We want to pull refugees from the waiting room only twice, once for registration, and once for signatures, fingerprints, and receiving documents.
- We wanted to improve the process of bringing refugees from their lodgings so we have to send back fewer refugees without registration due to lack of papers, etc.
- We made lots of little improvements, from a camera that zooms on faces automatically to having printers that handle two types of paper.
- We visualized using icons and multiple languages.
We are currently planning a new workshop for early 2016. Since the LEA is moving into a new building, the workshop content depends on the still unknown moving date.
- If the workshop is well timed with the move, we want to implement the new flow in the new offices, including visualization.
- Otherwise, we look at the boarding of refugees in the lodgings, including visualizations.
Our war room in the LEA didn’t have flip charts. I tried a new product here for the first time, and liked it quite a lot: Electrostatic flip chart foil. It is like a 60 centimeter wide and 20 meter long Saran Wrap in white or with squares. It clings to the wall, and you just cut off a desired length horizontally or vertically. Normal white board markers can be used on it. It is even dry-erasable, although this did not work well with a textured wallpaper underneath. The product is called Leitz EasyFlip, but at the time of writing it was unavailable on Amazon.com but available on Amazon.de. You can also see it in the war room image below on the left.
Overall, we had great fun and the warm fuzzy feeling of doing good. The entire workshop was done without the help of computers (except for e-mail communication). All ideas were developed on paper. A more detailed description of the processes (in German) can be found on Lean for Refugees. We are looking forward to the next workshops. I will post updates here too. In the meantime, go out and organize your industry!