|
|
|
10/06/2011 12:43 PM
|
|
|
Not sure if this is a usual "first post", but here goes...
I work in a large building with contractors, consultants and various other teams. We recently implemented a hot desking system (as opposed to a manual system of booking desks/facilities) and things aren't going well. People are supposed to book their desks in advance - and they forget. People end up not seated next to the rest of their team mates - they end up either on a different floor or in a different building. People are supposed to "unbook" their desks when they are sent out, e.g. to a customer site - but don't. So you can turn up and find no available desks on the system but there are clearly desks not being used. If you take a vacation, a colleague has to book you a seat or else you will be sitting in the car park on your return! And if I get called out of the office, I can't easily "give" my seat to someone else who has just turned up.
Now, this system was expensive and it is going to be difficult to persuade the Board to spend more money on something better. But, we are losing money because of the new admin overhead of booking, or rather mis-booking, somewhere to sit.
I am not used to using LEAN outside of a manufacturing system, but I think a LEAN perspective may help me convince the Board that this position isn't sustainable and needs changing. Can anyone suggest how to apply LEAN to this please?
|
|
|
|
|
10/07/2011 01:08 PM
|
|
|
How would you use Lean to solve a problem of resource allocation in manufacturing?
You may be too close, step back and have a look at a higher altitutde and then focus in on solving the problem, just as you would in manufacturiing. The lean toolbox is just that, a set of tool to use as needed to solve a problem, doesn't matter where the problem lies.
|
|
|
|
|
10/07/2011 05:23 PM
|
|
|
Not sure if this helps but...
Idea 1: I have seen this work quite well with a first come/first served approach. The teams were small and you could see almost the entire area so that as people came and went you could visually manage yourselves. No software, no reservations.
Idea 2: Stop having to un-book desks. If the reservation was for only a limited time (8 hours?) then open seats would show up in the system faster without an action required.
Idea 3: Reservations done through the network ports so that when you sit down and connect your computer the system understands that seat is occupied (because the port is being used). The system "pings" every so often (every 2 minutes for our desk phones, actually) so that when the port is no longer in use the seat is marked "open". Almost real-time and without the need for seperate action of booking and un-booking.
Booking is like forecasting and has all it's errors and inaccuracies.
Good luck.
|
|
|
|
|
10/10/2011 11:48 AM
|
|
|
Hi there, and thanks for the responses.
Some additional information - I should have mentioned that the building runs large teams who need to work together. This makes for difficulty when the manual system breaks down (they forget to book) and other staff book in the large team's location. They need to do this as the building is at capacity and every seat is needed. Worse, the teams have set locations so a free-for-all wouldn't (actually couldn't) work.
I am thinking that the definitive solution is auto-seat allocation based on whether a staff member is scheduled to be in or not. That way the 8 wastes are minimised. Fundamentally, paying people to book a seat (badly!) is poor business isn't it?!
|
|
|
|
|
10/11/2011 03:10 PM
|
|
|
Hi Grumpy,
1. Have you considered using the 7 wastes as a framework to analyze/measure/present the impact:- ?
Transportation:- time/distance moving the "product" (e.g. the information, tools, files people need for their jobs)
Inventory:- product piling up (has there been and increase in work in process ?)
Motion:- time/distance moving the people (e.g. to contact their team members, fetch files, ...)
Waiting:- time waiting to get a desk
Overproduction:- is manual booking effort really required ? could it be automated from people's calendars ?
Overprocessing:- is the effort required to book a seat so complicated that people don't ?, could it be simplified ?
Defects:- has there been an increase in the error rates ?
2. It sounds as if your organization has taken the approach of buying a "system" (= tool) before having
mapped the value stream (process) , streamlined the process and/or trained the users on the process.
May be smart to do this in retrospect. If you find you have lots of team interaction in your processes, maybe the unit of booking
should be a team and its size not an individual.
3. You may be able to refine the process with e.g. demand levelling, error-proofing tools (for booking),
and get the expected value-add from the tool. The value-stream mapping might lead to segmenting the hot-desk resource into
"product" lines and "cells". Spaghetti diagrams might also be useful here. Also look at potential layout improvement
- does the system allocate from the edges towards the center (so free capacity clusters together) ?
4. You could use simple kanban/visual controls - e.g. if you want to give up your seat - hang a big green card on it on the way out,
at the end of the day, these cards are taken off.
Hope some of these ideas help,
Cheers,
Pete Senior
|
|
|
|
|
10/11/2011 03:10 PM
|
|
|
John,
It sounds like you have a very interesting problem. I have a number of thoughts on this, but part of developing a lean solution is to have a clear definition of the problem you are trying to solve. You mention that the buildings are at capacity. Why was the hot-desking initiated? Was it because team movement was too frequent for a planned seating layout? Or was it meant to be a method to simply squeeze in more people? I think you fundamentally have to decide to have a planned environment, or an adaptive one, and stick to that. The situation you describe seems to be trying to apply a planning methodology to an adaptive system; mixing the two is both common and rarely successful.
Auto-allocation sounds to me like MRP for seating. This is a good idea in theory, except for all the surprises that can happen. For example, are people called to customer locations at the last minute? Do teams change their membership dynamically? The more this happens, the less you will be able to plan you way out of it--you are adapting to changes in the situation, no matter what system you have.
It would seem to me that this is a similar situation to seating in a restaurant, although on a different scale. You may be able to learn a lot about your problem by observing this similar situation. If you look how most casual-dining restaurants operate in the US, they generally take / require reservations only for unusually large parties. If you partitioned your buildings into blocks of typcial party size, you may be more flexible. (so, a conference room for 10, or office area for 20 rather than a table for 2 or 4) Then, manage reservations for larger groups that have to be together, but have difficulty doing so on their own.
I would also agree with Bob about visibility: if you can't trust the reservations system, then you need a way to empower your people to understand what is free, and what is busy. Returning to the analogy of a restaurant, you have a host / hostess who "schedules" the tables, and adjusts for unforeseen issues. In this case, they need central visibility, and in return they give the scheduling orders. Another example would be a doctor's office, which has a lot more people involved in scheduling / pacing. Here, I have often seen small flags outside the room. In their case, there are many flags to designate different states, but a simple occupied / unoccupied flag may be sufficient. (and make sure it is visible enough that someone can scan a whole hallway, not have to walk by each one to find a free office) I also like Bob's idea of utilizing the LAN as a natural way to see if someone is there.
One fallacy in the restaurant analogy is that, at a restaurant, you also manage a waiting list. I assume people won't wait for a day for the next available desk. However, one other purpose of the waiting list is to encourage people to eat at different times--no waiting at 5:00, or at 9:00. If you are at capacity, has your management considered early / late hours for usage? It's quite common in manufacturing to free up utilization by addiing different shifts. (Understanding the personal cost needed to adjust, but perhaps a number of people would benefit from it, and volunteer)
|
|
|
|
|
10/12/2011 02:55 PM
|
|
|
This is a very interesting issue, indeed. I fully agree with Peter and Matthew.
|
|
|
|
|
10/17/2011 09:41 AM
|
|
|
Hi All,
So sorry, am abroad on a conference run and haven't been able to do justice to your kind responses. Bear with me! I'll be there soon!
Thanks again!
|
|
|
|
|
11/01/2011 07:36 PM
|
|
|
Grumpy, Best to start with defining the problem.
For example is un-used space below target? Is capacity below demand? Is there lost revenue (customers who leave becasue they don't like their assigned space, etc.)
Any discussion on a course of action to correct is pre-mature until a single over-arching problem is defined.
|
|
|
|
|
01/16/2012 10:16 AM
|
|
|
Why the need to hot desk in the first place? What problem did it solve?
What was the goal of the hot desking?
Assuming there are more people than desks, before the implementation of the new system were people fighting for a desk space?
It would be interesting for us to hear some background/history on the situation.
Cheers
David @ Leansecrets.co.uk
|
|
|
|
>
Community >
Forums
Send to a Friend
