Posted by: Knightbird | August 26, 2015

Alaska Division of Public Assistance

I have admired and written about the Alaska Division of Public Assistance before. The are pioneers in Lean Government, having started in 2009-2010 with a project I wrote about earlier. Since then, they have continued, but I heard very little about their progress. Today, I read an RFP for lean process consulting. It’s an admirable step. But it’s also a sign that perhaps they have not found true Lean. Because I do not have the capacity to bid on the project, let me speculate about the genesis for the need.

Because DPS is a large division underneath an even larger department that I have been interacting with for over 3 decades, I do know that DPA’s initiation of Lean is not a Department wide initiative. This tells me that there is no “strategic” leadership. The benefits of Lean at DPA were probably tools based. A kaizen is conducted through which amazing improvements emerge, but are not necessarily sustained. With 500 employees, they should have a cadre of Lean Specialists trained and conducting improvement events regularly.  I don’t have enough data to propose a right sized work force, but I would not be surprised if they have 20% more than a truly lean organization would need.

A second clue to the lack of strategic lean focus is the flawed implementation of their benefits management system. Referred to as “Alaska’s Resource for Integrated Eligibility Services” (ARIES), it appears that in their haste to implement an automated system, they violated one of the central tenet’s of Lean. Don’t automate unless there is a compelling reason to do so. I haven’t read of a compelling reason, other than certain requirements imposed by the federal government. It’s a fact of life that IT projects are expensive and considerably flawed in almost all implementation. The Municipality of Anchorage has a severely flawed implementation of its ERP software package. At an initial cost of $10 million, it has ballooned to over $50 million.  The new mayor has apparently put the project on hold, but in the meantime, a huge resource has been bled from public coffers, with a substantial price. $50 million invested at a 7% annual rate of return could generate $3.5 million in annual revenue. In order to do a fair comparison for evaluating the benefits of automation, we would need to have a highly efficient process costed out. At this stage in DPA’s Lean evolution, they acknowledge they had a backlog of 10,000 applications.

At this point, I want to make an obvious observation for a Lean Strategic System. Because there is not a strategic Lean vision, there are ineffective “Pull” systems in place that extend beyond DPA. I envision a lot of batching in supportive systems throughout DPA’s area of expertise. I am sure there are federal pull systems that hinder their operations substantially as well, since they manage federally funded programs.

From the information provided by DPA, we see that they serve 150,000 Alaskans monthly with a staff of 500 located in 11 offices. This isn’t enough data to speculate on where more efficiencies can be gained. Using some rough analytics, 150,000 total clients average out to 300 clients per staff per month. Because there are various processes, all I can speculate about is that they have a daily Takt time of 15.1 Clients per minute (150,000/22 days=6,800 per day; 6/800/450 minutes=15.11). Cycle time isn’t obvious because no data is given. We can assume given the wide range of programs administered that the tasks involved are of varying degrees of complexity I can speculate that we need to load level (heijunka).

So what would my recommendations be? First, I would try to understand the current system by mapping it out. I do know that they have a minimum of 8 programs they administer. Each applicant must meet eligibility requirements, I would guess for multiple programs. I would try to identify the wait time within each process, but basically, there is one entry portal with multiple paths to travel. We need to identify a cycle time for each pathway. We would need to identify the documents used and the pull systems required for the full cycle. Usually the waiting time generated by inefficient pull systems greatly adds to the lead time for a cycle. Any good measure of a process also includes the defect rate and any feedback loops generated. Feedback loops area huge time waste that can be eliminated through standardized responses.

The actual improvements can be generated in one week with a series of kaizen participants working on different parts of the processes. We have to just mitigate the pull systems the best we can, but for our processes, we want to do the following.

First, the intake/application form can be standardized for each program with a single Personal Identification Section. (PI). Questions answered lead to additional sections. For some programs, we need to know if there are dependents and their status. Regulations define the extent to which we need to collect and verify this data. So our questions are linked to the regulation and form the basis for audit, if needed. We did this exact same process when we did a Kaizen for our Head Start program at a prior employer. Our file pattern was linked to the regulations and allowed us to verify required information within the file. Any gaps in the file pattern meant we were out of compliance. What we want to do is gather the appropriate information, and no more. Further requirements depend on answers to previous questions and if the data is not required, it is not collected. The application process is mistake proofed to our best ability. We want a complete application when it arrives. The application should then flow as best it can through a process that leads to eligibility determinations and action. We reduce or eliminate as many waiting periods as we can. We reduce or eliminate feedback loops and defects. At the end of our Kaizen, we have implemented as many improvements as we can and at the end of the week, we are ready to train staff on the process. We roll it out and measure the new process for improvements. At the same time, we empower staff to resolve defects at the time they occur.

Ultimately, I would expect about a 20% reduction in staff time required to complete the work necessary to clear 15.1 clients actions per minute. With 500 employees, there is 33 minutes available every day of total possible value adding time per client.

DPA is to be commended for their continuous quest for lean knowledge. The should be supported by their Governor, Commissioner, Deputy Commissioner and Director, working as a team. When the state figures it out, I expect DPA will become the shining Lean Start in our state portfolio of services.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: