iSixSigma

Another time study question

Six Sigma – iSixSigma Forums Old Forums General Another time study question

Viewing 33 posts - 1 through 33 (of 33 total)
  • Author
    Posts
  • #42385

    IE
    Participant

    I’m an IE for an assembly line in my plant and we currently have no data to represent any of the processes on the line.  After working out on the floor at the various stations and observing the flow of WIP, I’ve seen where the flow slows; it slows at a T-junction where two pieces of WIP are combined to make the final product.  What’s the next step?  My first instinct is to do time studies and look at how long it takes to make WIP A, WIP B, and how long it takes to combine the two WIP’s to make a final product.  It gets more complex because there are several varieties of products, and occassionally we run into quality problems that we don’t catch until the final assembly.  Where would you guys recommend I start?
    Its funny that I’ve got my degree, but I still have trouble approaching problems like this.  It makes me question my ability.  Is this something I’ll just pick up over time with increasing experience?  Even my colleagues and peers don’t seem to know how to approach these big picture problems.
    Thanks, Andy

    0
    #133757

    OLD
    Participant

    IE:
     
    Step #1 – Throw away your stopwatch
     
    Step #2 – Read “The Goal” by Goldratt
     
    http://www.amazon.com/gp/product/0884270610/104-7032498-6088766?v=glance&n=283155
     
     
    Step #3 – Define: What are you trying to accomplish?
     
    Step #4 – When you know what you want to accomplish – identify tools that will help to those ends.
     
    If you can define what you’re trying to accomplish, many people on this forum can help get you started by suggesting the tools. The tools will become more obvious to you as you gain experience and knowledge.
     Good Luck!  OLD

    0
    #133758

    Peppe
    Participant

    I’m just curious to know how production manager, manage the production.
    Rgs, Peppe

    0
    #133759

    IE
    Participant

    Its funny you should mention that, I’ve read the Goal twice and It’s Not Luck as well.  I understand the theory, but I never have been able to apply it; its just always been more complicated when I see what actually happens on the floor.  I’ve asked questions on this forum about the Goal before, but they’ve always been unorganized questions.  Now I think I have it narrowed down to two simple questions:
    1. What do I do when the bottleneck moves around due to product variation and rejects/reworks?
    2. If the theory of constraints is such a great thing, why isn’t that all industrial engineers spend their time focusing on?  For instance, send me out on the floor and I’ll just constantly work on improving the bottleneck, then improving the next bottleneck, etc.
    I find myself falling into the trap of not defining the problem and jumping to conclusions.  Everyone I work with does the exact same thing.  I hope that over time I overcome this problem.
    Thanks, Andy

    0
    #133760

    EIEIO
    Participant

    *2 Sigma = .4 lambda
    *Lambda-Tau CalculationA calculation method used to determine availability or short-term reliability when the age of a component or system is unknown or indeterminate. Relex Fault Tree supports multiple methods of Lambda-Tau calculations. In all methods, Lambda signifies the failure rate of the system or component. Tau can represent the inspection interval, repair time, or mission time of the system. Takt* Time ( German for beat) = time available/customer’s demand

    0
    #133761

    IE
    Participant

    Peppe, maybe I should’ve mentioned this before but we “make to order”.  In other words, we don’t start producing a product until the customer has already submitted the order.
    The supervisor of the line doesn’t really follow the bottleneck like he should.  But maybe I’m not answering your question.
    Thanks, Andy

    0
    #133764

    Peppe
    Participant

    Maybe I wasn’t clear. My question is how your production manager is able to manage production not having detailed data about capacity, bottleneck, lead time, scrap, etc.. How your production line have been set up ? Based on what ?
    Rgs, Peppe

    0
    #133765

    OLD
    Participant

    IE:
     
    Step #1 = Toss the stop watch – Complete (please say that you did)
    Step #2 = Read “The Goal” – Complete
    Step #3 = Define what you want to accomplish
    Step #4 = Identify tools that will facilitate #3
     
    Have you completed Step #3? If yes, please restate to clarify.
     
    TOC is one of many tools. It may/or may not be the tool of choice for your situation. Complete #3 as best as you can and let’s see what advice you get from the forum as to the tools of choice…..
     Good Luck! OLD

    0
    #133768

    KKN
    Participant

    A couple of tools I would look into using would be a
    Value Add analysis (to reduce the amount of time spent on Non-Value added steps) and line balancing, using TAKT times.

    0
    #133769

    IE
    Participant

    There are so many things I want to accomplish on my line.  I’d like to improve quality and production(we will be facing much larger demands in the future…larger than even our current “good day” production numbers), but I would venture that those goals are too general to simply choose a tool.  I would guess the first step to accomplishing the two big goals is defining smaller goals.  For instance, I want to understand why there are so many problems at that T-junction in my line and then take corrective actions.  Is this a good approach?  Maybe I need to learn more about general problem solving?
    Thanks, Andy

    0
    #133770

    IE
    Participant

    My company is rather small and the closest thing to a production manager is probably the line supervisor.  Other than that it would be the plant manager.
    The supervisor manages production solely from intuition.  We have hardly any detailed data about capacity, bottleneck, lead time, or scrap.  He has almost no information other than current production for the day and what he sees around him.
    I’m not sure what the setup of the line was based on, but I know its supposed to be a “pull system”.  The biggest areas of WIP buffer inventories are at the T-junction.
    Thanks, Andy

    0
    #133771

    IE
    Participant

    I’ll look into value added steps, and I’m ashamed I haven’t thought about takt time.  Of course I don’t have cycle times to compare to the takt time either.
    On the subject of line balancing: this is one of those things that has always confused me, the Goal says that line balancing doesn’t work and is asinine.  Other tools like VSM include line balancing.  I just don’t understand that conflict.
    Thanks, Andy

    0
    #133774

    KKN
    Participant

    In my humble opinion, line balancing at a gross level should be a goal, but natural process variation makes it close to impossible without building in WIP buffer. As an example, I’m currently working a transactional project where step a has 2 people working and cycle time (touch and wait) is around 1.5 hours, and step b has 7 people working with it’s cycle time around 5 hours. While I agree that it would be silly to try to balance these steps down to a knat’s rear-end, but I am going to simplify and move some task from b to a to better balance the line.

    0
    #133775

    OLD
    Participant

    IE:
     
    Your enthusiasm is admirable! Attacking on too many fronts is probably not the best way to start. You are thinking wisely that you need to narrow your scope. Continue to try to be more specific. What do you want to accomplish or, what is causing your line the most pain?
     
    You mentioned 1). increasing production volume and 2). improving quality as two goals. Thinking practically, you may want to improve your “quality” before you increase your production. More volume without an improvement in quality means more rejects. Improved quality means less rejects (that also increases throughput).
     
    If you start with improving quality as your goal, what is happening on the line now that is creating rejects? Collect data:
    How many rejects?
    What is the rejection?
    Causes for rejection?
     
    Again – What are you trying to accomplish?
     
    In your answer to Peppe, you indicated a Line Supervisor and a Plant Manager as other players in your company. You may want to enlist their help in establishing your improvement goals? Start small – be specific…
     Good Luck!  OLD

    0
    #133785

    IE
    Participant

    What is causing my line the most pain?  I’m not sure which is the biggest contributor to poor overall performance, but I’d be inclined to say that instability is the biggest problem.  Generally this can be traced back to rejects/reworks.  Unfortunately I’m quite sure that if I asked others in the plant about the biggest contributor, they’d immediately come up with a quick answer without thinking about it; this wouldn’t help me much.
    Here’s an even better question: Is there anyway I can find out what the biggest pain on my line is other than pure intuition?
    Thanks, Andy
     

    0
    #133788

    IE
    Participant

    So, in short, line balancing is a good tool to use but the natural variation between stations will make sure you can only balance the line to a certain degree.  Does that sound right?
    Thanks, Andy

    0
    #133789

    villageidiot
    Member

    It appears as if you are looking to make inprovements before you have defined your efforts. You are dealing with informal processes with no measurment system in place.  I am assuming some prior training.  Here is one method:

    Organize all your SKU into product families.  Use a product family matrix (Learning To See).  This tells you what goes where and why.
    Pick a group of resources (line or cell) based on something  strategic:  If you are a small company who seeks to be a low cost provider, use an ABC approach to identify teh product family that has the greater costs or cost drivers. Etc.
    Map it. Include cycle times, inventory counts, headcount, ABC for each process step.
    Choose a segment of the line for your efforts based on your mapping results. Use a VSM event to reduce costs (reducing headcounts, handling, transport, mobility, inventory levels, etc.) or increase your process velocity (cycle time, productivity, capacity, etc).  Sigma would work well for defect reduction if the problem and solution are unknown. 

    Keep your efforts simple.  Sigma is probably too much firepower for your immature processes.  Seek standardization and lean efforts, as they are more user friendly, have faster implentation times, and will stabilize your process so that sigma tools will be more effective.  Good luck.

    0
    #133790

    OLD
    Participant

    IE:
     
    Pain = $’s. Ask the Line Supervisor, ask the Plant Manager, and ask the Accountant: what/where are the costs associated with manufacturing this product? If indeed, rework is causing instability in the production flow then try to quantify the cost of rework. How many units need rework? What is the cost of the rework in hours and dollars? If rework time takes time away from production, what is the lost opportunity? Follow the dollars….
     
    By asking the question of where are the cost $’s, you will begin to see opportunities for improvement. When you have identified the opportunity, establish an objective (reduce rework by 75%). At this stage, the tools to use will become clearer to you. You might be surprised by some of the fixes that will become obvious once you have exposed the opportunities.
     Good Luck! OLD

    0
    #133792

    KKN
    Participant

    Exactly

    0
    #133824

    Peppe
    Participant

    Further detailed good advices already received, I whish to add some very simple point. Looking at you production line as one big box (as first approach and then apply the same for every sub processes), you could have, in any case, at least two data on which you can start a basic analysis: purchasing dpt (or incoming dpt) and shipping dpt (or invoice dpt). Based on basic equation of  WIP= incoming-shpped and that any item shipped is composed by A=a+b+c, you can start to draw the materile flow, even if the management is by “intuition”. The time between different purchase of same raw parts cangive a sort of “lead time” about it.  The rate between purchaed items and their WIP can give you some basicaly indication. After that is basicaly that you start to define a process map and the time for each activity in production line. You can simply start with an average of  “qty shipped/time spent” and then go on for each sub-processes, as per advise received. About your question on theory of constraint, I think you received some good answer, anyway if you have further specific question, please let me know.   
    Hope this help.
    Rgs, Peppe

    0
    #133832

    Johnny Guilherme
    Participant

    Andy Hi i am not quiet sure what is your objective in what you have posted. But if i understand you correctly then what i suggest is that you define each of the processes per the variaty of products you have. Then you will have to do a time study on each of the processes i.e. time each process over a good period of time and try to rate the worker doing the process and obviously get a time per process. Using a typical stopwatch with a TMU units could do the trick. You can see someone working at snail pace as apposed to someone puting in some effort and working deligently. Once you have each time per process per variant, then you can understand the utilization at each process/person, you can define the bottleneck operation and you can also then understand how many units could be made in the time available. This helps you set up a standard for the line.
    You can then use this standard for capacity planning as well as cost purposes. With regards WIP-people will always build up WIP, because they are comfortable having WIP around. WIP can be romoved once people are working in a synchronised fashion.
    I am not sure if i have answered your posting since you have not clearly defined what your objective is. I have a line standard sheet that i used to use when doing time studies and trying to balance lines. If you want it let me know.
    regards
    Johnny

    0
    #133833

    IE
    Participant

    I think I can find the value of the material for a rework/reject, but I’ll have a hard time quantifying things like lost production opportunities because I have no data on cycle times; e.g. what the reworker would have produced had he/she not been busy reworking a unit.
    I’d guess that I’ll find the dollar value of the reject/rework itself won’t be nearly as significant as the time and effort people put into taking care of rejects/reworks.
    As far as collecting the data goes, I believe it would be impractical for me to stand out on the floor for 8 hours because there are generally about 4-8 rejects/reworks a day(when we generally produce about 60 units a day).  Any recommendations on how to get accurate data from the floor?
    Thanks again, Andy

    0
    #133862

    OLD
    Participant

    IE:
     
    Who finds and/or determines there is a reject? That person could be of help in collecting the data. Other options for gathering data:

    You could collect the data and it probably wouldn’t be a waste of your time as you could track each reject upstream to the source. Seeing first hand what is going on and when, may be of more value than your time spent elsewhere?
    You could ask an operator/assembler to collect the data
    The person that actually does the rework could collect the data
    You could ask the Line Supervisor to collect the data
     Start simple but start. 4/60 = 7%, 8/60 = 14%. Make some assumptions to quantify your costs…..
     Good Luck!  OLD

    0
    #133895

    IE
    Participant

    Allright, then it sounds like I’ve got a place to start.  Thanks for the advice/guidance.  I hope in the future I won’t need advice though.
    Thanks, Andy

    0
    #133898

    IE
    Participant

    I just had another thought!!!
    I was thinking about improving quality and its quite possible that I’ll be using Lean tools to accomplish my goals.  For instance, I think standardization and poke yoke will be used extensively.  My question is: Should I use these tools now?  Or should I use them during the course of value stream mapping?  This has often popped into my head, and its made me step back from projects in the past to question their priority.  What do you think?
    Thanks, Andy

    0
    #133899

    OLD
    Participant

    IE/Andy:
     
    Cool! Take a whack at it. It will be a learning experience and this experience will bring knowledge to be applied to the next problem.
     
    Good Luck in your efforts! OLD
     PS. Asking questions and seeking advice are good things. Combine that with your own efforts of researching/trying/doing/making mistakes/etc., and you will get closer to where you want to be. We never stop learning!

    0
    #133907

    OLD
    Participant

    IE/Andy:
     
    This thread started with your question pertaining to a bottleneck at the T-junction. You indicated that one of the reasons the T- junction is a bottleneck was due to the rejects/rework. Start there and focus on the 4 to 8 rejects that are happening each day/shift

    Collect data as to why/how they are failing
    Pareto the reasons (most likely the 80/20 rule will apply)
    Look for Quick Wins
    Make the changes
    Monitor the results
    Revisit/Control/Adjust as needed
    Identify the next contributor to the T-junction bottleneck
     
    No, you are not done when you complete your first pass at the rejects and Yes, it is a good time to use those tools you have mentioned. This is an opinion based on too little information (and thus, likely to be wrong) but it appears you are attempting to “flock shoot” to gain improvements as opposed to targeting specific pain areas? From what you’ve mentioned in your posts, your company may need:

    More leadership/involvement from the Line Supervisor, Plant Manager, and others
    A strategic and/or philosophical approach to continuous improvement
    Basic measurements, controls, and procedures to support the production efforts
    Other???
    Attempt to resolve what you can control….
     Good Luck! OLD

    0
    #134017

    IE
    Participant

    Thanks again for the reply.  I think we’re on the same page.
    I’ve been thinking about this more and more recently and I think I may be asking more of an ethical question.  What I mean is: If I do value stream mapping, I’d have a more formal recording of the improvements I’ve made, and hopefully the company would think much higher of me and the job I’m doing.  By value stream mapping the line, there would be a much longer delay before I actually made improvements, thus hurting the line.  If I strictly focused on reducing reworks, I would be helping the line much sooner, but it wouldn’t necessarily appear that I made as much of an improvement.  In retrospect, I think I can compromise by simply improving the reject/rework data we collect and how we collect it.  I know that by considering my “image” I sound self-centered, but I’m just being honest.
    I’d like to stabilize the line in terms of scrap/rejects/reworks before looking at production improvement tecniques(although I understand that eliminating rejects will naturally increase production somewhat) like VSM.
    Thanks, Andy

    0
    #134378

    R.M.Parkhi
    Participant

    You should start with two products, since you have a no. of  products running in the same assembly line. They are:
    1. Which has highest volume of production,
    2.Which has the highest quality problems.
    After this, you should do detailed time study analysis of each & every process.Pl. try to balance the timing for all the stations so as to enable you to determine ‘ Takt Time ‘.
    This activities will open the window for further actions.While studying the process time ,you may notice certain operations taking unusual floor to floor time due to quality problems.Try to tackle them first.This way by following Pareto,you should be in a good position in near future.
    Every I.E. passes through similar difficulties in the beginnig of his career.
    With best wishes,
    R.M.Parkhi

    0
    #143423

    Israel
    Participant

    i think the best way is to define first the problems.
    second you standardize all the things that happen in the workplace.
    make a flow process chart, gang chart and man machine chart to evaluate the whole process and to find other ways to improve the process of your department.
    when you done all this things you need to re asses the whole process so that you can compare what went wrong, what are the things to improve, and what are the precosunary mesures that needed to mentain what is good…

    0
    #148564

    KSU IE
    Participant

    Andy,
    I know this has been out there for some time, but I wanted to give you some reassurance.  You are applying the priciples in “The Goal” when you looked at the production line and saw all the WIP in front of the T-Junction!  Your bottlenecks will always have the most WIP and those operations downstream will be starved.  Now you need to determine how to improve the bottleneck (modify or split the process to get more production) or, if not possible, combine operations before and after to match the pace of the bottleneck (no need to build parts to just wait for the next operation).  Then use the workers no longer needed on that line to start another line for other production.  As you’ve seen, long assembly lines with a wide variety of product is difficult to manage – many smaller, dedicated, lines work better.  This is the cell concept and fits with Lean.  This allows the right parts to be staged at the same location and the same workers to gain skill at building a few products well.
    As for looking good in the eyes of management – managers may pat you on the back for creating charts but the company will profit by reduced waste and increased capacity.  This in turn will help keep your company competitive and you in a job.  You can work on the overall process map as you conduct studies in each area chasing that bottleneck.
    Good luck is usually brought about by perseverance.

    0
    #148997

    KSU IE
    Participant

    Andy,
    I know this has been out there for some time, but I wanted to give you some reassurance.  You are applying the priciples in “The Goal” when you looked at the production line and saw all the WIP in front of the T-Junction!  Your bottlenecks will always have the most WIP and those operations downstream will be starved.  Now you need to determine how to improve the bottleneck (modify or split the process to get more production) or, if not possible, combine operations before and after to match the pace of the bottleneck (no need to build parts to just wait for the next operation).  Then use the workers no longer needed on that line to start another line for other production.  As you’ve seen, long assembly lines with a wide variety of product is difficult to manage – many smaller, dedicated, lines work better.  This is the cell concept and fits with Lean.  This allows the right parts to be staged at the same location and the same workers to gain skill at building a few products well.
    As for looking good in the eyes of management – managers may pat you on the back for creating charts but the company will profit by reduced waste and increased capacity.  This in turn will help keep your company competitive and you in a job.  You can work on the overall process map as you conduct studies in each area chasing that bottleneck.
    Good luck is usually brought about by perseverance.

    0
    #183931

    Plant I.E
    Participant

    0
Viewing 33 posts - 1 through 33 (of 33 total)

The forum ‘General’ is closed to new topics and replies.