Work Measurement PFC & Time Study Webinar

[Audio] Hello everyone, and welcome to the global Industrial Engineering webinar series. The Global IE team is presenting a series of 7 LIVE webinars to re-enforce foundational IE practices to all engineers of all levels; and, to introduce this as new material to our new engineers as well. As a reminder, all presentations will be given in English… we kindly ask you to convey this information to those that will need it in your local languages. My name is (SAY YOUR NAME) and will be your presenter today. The experts onboard are (LIST THEIR NAMES). They will help answering any question you may have during the presentation by using the Q&A chatbox in the upper right corner of your screen. Our LIVE Q&A session will take on additional questions right after the presentation, and we will feature several questions so all may see and hear our response. This LIVE event is being recorded! The recording and powerpoint slides will be available in about a week in the IE Portal. In the unfortunate event of a network connection issue, you will still be able to download the recording and watch it. Additionally, all of the Q&A questions will be recorded as well. Today we will focus on Introduction to In-Process Kanban Webinar… thank you for joining, and we will now begin. CLICK TO NEXT SLIDE.

[Audio] The content that we will cover in this webinar are as followed: Definition Objectives Why we need In Process Kanban In Process Kanban Types Setting Up In Process Kanban Limit Special Cases/ Example Summary.

[Audio] What are In Process Kanban? or IPKs, for short IPK is the amount of Work In Progress (or WIP) within a continuously flowing line, between different manufacturing lines, or within a line that requires a batching process. WIP are units that are physically being built at a workstation, or units that are waiting to be built by the next operation or manufacturing line. IPKs are a core lean property to limit WIP in our value stream. They are key line design tools that allow us to create smooth workflows which enable us to optimize the performance of our production lines and prevent WIP overload on our manufacturing floors..

[Audio] Production processes with good IPKs designed into them will have good effective line management because: There will be good visual control to help prevent WIP overload We can identify quality defects sooner because IPKs will streamline WIP IPKs help alleviate bottlenecks in our workflow in realtime…they Keep Bottlenecks stations running continuously by alleviating process imbalances, natural human variation between stations, and alleviate the impact of line interruptions They Ensure minimal operator downtime due to no WIP. Keep production lines neat and tidy with good 5S practices And ultimately, lines designed with IPKs, Streamline our value add to our customers. The whole idea of setting up In Process Kanban is to create a smooth and healthy flow to maximize our production resources, both operator and machine resources. Therefore, our system will create value more efficiently. By effectively designing visual control of WIP entering and leaving our production lines, we create visually effective management of our lines with in-process kanban..

[Audio] LEAN promotes one piece flow, so why do we still need IPKs? Due to the nature of our contract manufacturing business, especially in High Mix Low Volume, we need IPKs to make our production flow more efficient. Here is an example of a common manufacturing line with one piece continuously flowing through the line and no IPKs designed into it… This line could be made of manual assembly stations with machines, or all manual assembly stations. If we practice strict one piece flow with no IPK designed into it, then all stations are running at the bottleneck station cycle time… in other words, all of the stations are bottleneck time of 51s in this example…. Why? For strict one piece flow, the faster stations cannot build any WIP until they receive a unit from the previous process. Therefore, the operators will have high idle time because after they complete their unit they must wait for the bottleneck station to complete the next unit. Manufacturing lines will have natural process imbalances, human variations, and normal line interruptions, and IPKs can help… Not every line can be perfectly balanced with respect to time… there will always small differences of time in between stations Equally important are the natural Human Variations to each stations cycle time when building each unit Lastly, IPKs will help account for small line interruptions such as an operator talking to the supervisor, or using the restroom…. Even if that operator is not at the bottleneck station..

[Audio] In reality, in a continuously flow line, if we do not mitigate the impact of human variations and normal imbalances, what would happen? The realtime Bottleneck could jump around, WIP would begin to pile up between workstations, and workstations could become overloaded… All this might lead to: Quality issues going unnoticed, or not be detected as fast because units will not efficiently reach quality control stations due to WIP build ups More visually difficulty to manage the line Physical space will get constraint and congested This may cause Operators to become distracted and stressed, thus performance will drop; And the possibility of these factors could even cause a safety incident.

[Audio] There are 3 common types of IPK which we will talk about today. IPK design in between workstations, more specifically, it is the WIP waiting in between the workstations of a continuously flowing line IPK between manufacturing lines, which is also referred to as a supermarket. WIP is built up between manufacturing lines due to large differences in their cycle time outputs, especially between SMT and backend workcells. And the third is a Batch Processing IPK, for example ESS Ovens, Curing processes, Solo wash process, etc. There could be many IPK types, but the focus for this module will be on the above. Any other IPK type will likely follow the principles of these..

[Audio] Lets take a look at some of the In Process Kanban Limit Calculation Fundamentals: The number of In Process Kanban defines the performance of a pull system… in other words, how well we design control of WIP entering and exiting our manufacturing lines. If you use too few, we will have constant problems with interruptions, missed deliveries or idle operators and processes. If you use too many, then we will waste physical space, either on our floor or workstations, and also money on inventory. Therefore, in In Process Kanban calculations, we need to correctly identify factors that will impact our production lines. Lets review some of these factors in the next slide..

[Audio] What Factors Influence In Process Kanban? There can several. The first is Line Balancing, if we do not do line balancing then the cycle time variances will be large. In fact, properly implementing a good line balance is a pre-requisite before implementing IPKs. Machine/Tester/Equipment impact such as multiple resources feeding fewer resources (or vice versa). Also, if we are constrained by the number of available of machines, for example, our customers may only provide us 'x' number of testers to meet demand. Cycle Time, we have different cycle time imbalances between processes, and also imbalances between production lines.. Layout space for availability for supermarket IPKs. PF&D and operator variation: each operator will have different skillsets, the fatigue rates are different, toilet breaks are at different times. And, even the same operator will have natural cycle time differences building the same product. Workstation Size: Material availability and the layout at workstation to allow WIP to wait, and finally Physical size of units, either being too large or too small..

[Audio] In order to have an effective In Process Kanban, first we need to physically define the workstations, set the work method, define the standard cycle time for each process, and to execute to improve the line balance. As a reminder, lets take a quick look into line balancing. Line Balancing is leveling the workload between operations in a process. We need to balance the line to minimize the losses of time due to imbalances between workstations and to Optimize direct labor and equipment to minimize overall manufacturing costs. This graph is taken from the Line Balancing module. In the graph before line balancing you can see that there is a lot of variation between the cycle time assigned to each process. In the graph after line balancing you can see that the Line Balance was improved and cycle times are more level for each process and operator..

At least one IPK in between each workstation – small or large units.

[Audio] Lets discuss the 5 minutes of variation time… This is to allow the operators to continue work while there are normal process imbalances, operator variations, and small line interruptions. With the IPK in the line, we will be able to keep the line running for about 5 minutes in theory. What if we reduce the variation time to less than 5 min? Too little WIP to keep the line running until some mitigation can be done Approaching strict 1-piece flow, and increase the risk of the line immediately stopping And Will cause all of the line operators to be idle, we would be under utilizing our resources). If we increase the variation time to higher than 5 min? There will potentially be Too much WIP in the line and will be congested Slower detection of quality defects downstream Workstation space constraints, and be difficult to visually manage As an important note, this 5 minutes is a guideline - and work cells can adjust to their specific needs based on physical size of the units, operator break times, and shorter process times, etc..

[Audio] So Why do we need a higher IPK amount before and after the bottleneck Workstations? Lets take a look at some scenarios here. To note, we have changed the MA1 Cycle time to 33 sec to have a more clear explanation. Scenario 1: If we do not have IPK and MA1 operator is going for a short break, the whole line will stop. We can make arrangements to get a floater to replace the operator at MA1, but that will still take time to make it happen; and, line losses will occur. Now lets take a look at Scenario 2 that has IPK designed into the line… We put an IPK in with the limit of 8 before and after the bottleneck stations, and then 1 IPK for the rest of the stations. 8 units of WIP will eventually build up between MA1 and MA2, as the first station is faster than MA2. When MA1 operator is away, while waiting for the floater to replace them, MA2 can continue to run as normal, therefore our bottleneck stations continue to run without any interruptions for the next 5 minutes. However, what if we do not have any floater to replace the MA1 operator? we can arrange MA3 operator to replace MA1. By doing this we shall build up the IPK between TEST and MA3 station. We may not be able to eliminate all the losses, but we can help to reduce them… as you can see, operators and line leaders must have a defined work standard on how to account for another operator being absent – or another playbook being pre-defined. An Important Note : Any delays at Bottleneck Workstation is a LOSS we will not be able to recover. That is why we need to keep the bottleneck stations running..

[Audio] IPKs between manufacturing lines, or also known as a Supermarket between manufacturing lines. In Jabil, we have 4 common types of manufacturing lines which are SMT, Through Hole, ICT and Backend; depending on your unique workcell's there could be more or less types of manufacturing lines. What are the main factors to why we must design the IPK, or supermarkets, between manufacturing lines? Unbalanced UPH's between lines – will give us different throughput for every line depending on the assembly and process type Multiple resources feeding fewer resources (or vice versa) – for example, how many SMT lines are feeding into Through Hole lines; and how many Through Hole lines feeding to ICT lines? MFG Line FPY and Efficiency – It is also a risk keeping high IPKs, or high WIP amounts, as there might be upstream quality issue, therefore we might have excessive rework & scrap downstream if we allow WIP to become uncontrolled. High mix environments – with the challenges of varying throughput, we need well defined supermarkets to have a smooth flow..

[Audio] To be aligned with Material Replenishment concepts…. On our manufacturing floors, we commonly have a push system and a pull system throughout the floor. SMT is ran off of a forecasted schedule, and thus, pushes the WIP to queue up in a supermarket – this happens regardless if the Workcells can accept the work or not. However, our Workcells have a set demand, aligning more with our weekly customer demand, thus, they pull WIP as needed. X hours is our Run Time Guide… we will highlight this run time concept in the next slide....

[Audio] We will keep the same priority as IPK between workstations, which is to keep the bottleneck process running. In this case, it will be our bottleneck manufacturing line instead of an individual process. We first set the run time guide at 6 hours… again, this is guideline based on the demand, business, and resource constraints of our workcell… we will discuss this more in the next slide. So to get the IPK Supermarket, we use UPH Delta x Run Time Guide (Hour). The UPH Delta is simply the difference of the higher UPH minus the lower UPH of two different manufacturing lines. In this case, the difference between SMT and Through Hole is 68. So for IPK between SMT and Through Hole, we take the difference of UPH and multiply it by the run time guide. = (SMT UPH – TH UPH) x 6 = (180 – 112) x 6 = 68 x 6 = 408 This may seem a bit high, but remember, this is a mix of many products. Additionally, we are at the PCBA level, thus many units can easily fit in a magazine or a rack trolley. If we calculate in between Through Hole and ICT the IPK amount is 258; and between ICT and Back End it is 174..

[Audio] Run Time Guide of 6 hours allow each of the manufacturing line to run continuously, as each line and assembly has different UPH's and support a different number of downstream manufacturing lines. This will also help to reduce the losses when there are line imbalances & delays due unplanned downtime… Important Constraints to Design IPK between Manufacturing Lines are as followed: Total number of SMT, TH,ICT and Backend line where multiple resources feeding fewer resources (or vice versa) Physical space available in the layout; remember we need a strategic location to store our IPK amounts in a supermarket The different UPH imbalances between the lines And also the product mix that is shared on each line… this will compound calculations to a weighted average of UPH and demand. Common Solutions: Run Time Guideline is set to half of a shift length. For this example, we are assuming a 12 hour shift… our workcell is able to adjust this guideline according to our shift length, number of shifts, and other business needs..

[Audio] The final category for IPKs in this module is for batch processes… there are a few different situations where we must consider sending our batch sizes in defined IPK amounts… Often we need to prepare a batch into limited capacity processes with long cycle times…. For example, an ESS Oven running for 12 hours… this situation is quite strait forward, if we have an ESS Oven that can fit 20 units… then our IPK batch is 20 units before and after the ESS Oven. In-line Batching processes with higher capacity limits must match to bottleneck time of the line For example, a Small manually loaded oven, air curing time for RTV and epoxy, etc)… in other words, we need to allow space and/or time for our IPK amount to accumulate… we will give an example of this in the next slide. Design batches into the quantity that in a packout box, or pallet size if it is a larger unit. It is recommended to execute this strategy after Through Hole and ICT lines. Don't forget to consider BE yields which may slightly increase your IPK amount!.

[Audio] We need to make some assumptions for the example on how to calculate the Batch IPK… Oven has high capacity (meaning we will not be able to fill it up) Manual loading/unloading Cure Time = 30 min (1800 sec) From here, the Batch Size is equal to the process Curing Time divided by the line Bottleneck Cycle Time (not including the oven time). = this is 1800 seconds divided by MA2's cycle time of 55 seconds, the line bottleneck.. After solving, we will round up to get an IPK batch size of 33 units NOTE: The Same concept applies for air curing, just more physical space may be needed (i.e. units on trolley racks).

Assumptions: Daily demand 50k, 20 hr TAKT = 1.44s CNC 3 CT 100s, CNC 4 CT 50s.

[Audio] It is very important to set IPK in our line designs, but not all situations require an IPK to be designed. Lets take a look at some special cases where we only need ONE IPK, or do NOT need any IPK Very Long cycle times, for example, if the cycle time is 1 hour or more then we may not need any IPK. If needed, just one IPK amount at the most if there are several stations in line. Big Chassis builds, from the physical size and how much floor space a single large chassis build can take, you may not have enough physical space even for 1 IPK. For example, a large server cabinet build in the Back End Process. Often, these large builds also have very long cycle times. Very low volume, as an example, if the volume is 100 unit or less per quarter than, IPK may not need to be considered; or, at the most, just 1 IPK. Fully automated lines (SMT), if the processes and machines are stable with less human interactions; normally in SMT there is a small conveyor which allows for a standard of 1 IPK. NPI, work standards are not defined yet so we are not able to compute the IPK..

[Audio] Lets do a quick summary on why we design in-process kanban, or controlled WIP limits… first, let us considers what happens when we do not limit the WIP… The Workstation can become over loaded with WIP, Which can cause Quality issues to be unnoticed for a longer period of time. There is Higher operator idle time due to no WIP And the Operator might be distracted and stress Which could possible cause Accidents/Safety Incident What Happens When We Design an IPK to limit the WIP entering the system? It tell us where the real time bottleneck is occurring And we can Reduce our Quality Issues much faster The Workstation is clean and maintain good 5S practices To help Prevent accidents/Safety Incidents IPKs allow for Better space management and More effective visual line management.

[Audio] And This concludes the module, where we defined that in-process Kanban are essentially controlled WIP amounts, and there are three main IPK types: IPK between work stations of a continuously flowing line, IPK in between different types of productions lines, and batch processing IPKs. All of these are key line design elements to ensure our productions lines effectively & continually use our operator and equipment resources. But prior to setting an IPK there are pre-requisites before designing the IPK limit, our line should be balanced and have a set standard work and standard cycle time. We need IPK's due to normal process imbalances between workstations and production lines, natural variation per unit per operator, and various small line interruptions. A carefully calculated IPK will allow a limited WIP build up, but not so much that noticing defects go unnoticed or too much physical space is taken up. Remember, 5 minutes as a guideline for in-line IPKs, and half a shift guideline for between different production line types to create Supermarket IPKs. Lastly, we noted how not every process needs an IPK… some situations allow for only 1 IPK, and sometime none..

[Audio] Before leaving, we would like to thank you for your participation, and remind you that all the webinars will become available on the IE Portal. More detailed IE training modules will be available through Workday..