DMAIC Phase Three - Analyse

Brainstorming

I completed a brainstorming session with the operators to gain their input as to the opportunities they could identify and to engage them in the project. I conducted the session on the manufacturing line to keep the operators within their daily work environment and therefore within their comfort zone. At the start of the session I explained that the purpose of the session was to identify waste on the line and/or opportunities to remove waste. The operators were slow to provide ideas at the beginning of the session however when I drew on some of the more confident members of the group, all of the others realised that they too had valuable input relating to the workstations they worked on and before long the task at hand turned from trying to draw out ideas, to managing the group so that all their ideas could get posted. After the session I drew up a list of the ideas and set up a meeting with the Line Team Leader, Manufacturing Engineer & Quality Engineer to review which of the ideas could be implemented within the time-line of the project, the ideas that may require a longer time-line to implement & the ideas that were not feasible to implement. I then met with the operators again to provide a feedback loop, informing them of the status & future plans for each of the ideas. The brainstorming session provided some excellent ideas as to what opportunities were open to us immediately and in the longer term and was critical to ensuring that the operators felt empowered within the project.

 

Takt Time.

I completed an analysis of Takt time to ascertain the frequency at which we needed to build one unit, in order to meet revised customer demand, this would tell us the required heart-beat of the line.

Existing Takt Time
Takt Time =	Net Available Time		450 min/day X 16 shifts + 200 min/day X 5 shifts		492000		547	Sec. Takt Time
	Customer Demand		900 units/month		900
Required Takt Time
Takt Time =	Net Available Time		450 min/day X 16 shifts + 200 min/day X 5 shifts		492000		364	Sec. Takt Time
	Customer Demand		1350 units/month		1350

Labour by workstation "Bar Graph".

In order to analyse the information I broke down the time needed for each operator to complete one unit through each process step and placed it onto a bar chart. This provides a visual representation of the work to be completed by each operator. I included a continuous black line to signify the takt time & a broken line to signify the takt time when the yield of the process is taken into account. If any of the bars rise above the broken line it means that that workstation contains too much work content to be able to meet the monthly build plan requirement. I also colour coded each bar with batch work orange, machine cycles green; load/unload machines black & manual work blue. In colour coding these activities I can see if an operator has a lot of batch work to complete at a start of a batch which will result in a "first boat effect" at the start of each batch. I can see if an operator is being constrained by a machine cycle time and whether or not they are working or idle during a machine cycle. I can easily see how well balanced the work is between each of the operators workstations. In an ideal world each bar would be identical in height, meaning that all operators had exactly the same amount of work content to complete. It is critical to be able to identify each of these situations in order to identify and therefore eliminate waste.

The main activities that I've completed in the analysis are:

1.      Excessive batch work at many stations - I propose to remove as much of this work as possible from individual workstations and have these activities completed by one operator.

2.      Balance the workload - some operators on the line complete as little as 33% of the work completed by their co-workers. As part of my analysis I have identified a much more level balance of work between each of the operators/workstations.

                         Before Analysis                                                                    After Analysis

Here we can see by the graph that the distance between the top of the bars and the black line above them is greatly reduced in the after analysis graph. This means that waiting time or "fresh air" as I call it has been reduced. We can also note from the graphs that there are 13 bars/operators in the before, the same work is divided between 12 bars/operators in the after graph resulting in a reduction in the headcount required to staff the line. We can also see that much of the batch work (orange) has been removed for individual work stations and given to an individual workstation who will complete much of the prep work required for individual workstations.

This task was relatively straight for as I had completed the time studies myself and as a result was knowledgeable on the process and how it could be moved and divided to re-balance the work content.

WIP Analysis using "Little's Law".

 I used Little's Law which is the mathematical relationship (Throughput time = Work In Process X Cycle Time) to look at controlling the amount of WIP at each work station.

Standard Work.

In order to implement Standard Work I used a software package called Timer Pro Professional to analyse the following;

1. Sequence; a) the sequence in which tasks are performed & b) the sequence in which units are processed through each step of the work-station.
2. Inventory; how many units are present at a workstation & what is the minimum inventory used to keep the line flowing.
3. Layout,; is the line layout & station layout optimised for efficient & safe work practises.
4. Time/Technique; are the techniques used by the operators standard and adequately documented in process work instructions so as to ensure the standard can be maintained and improved on. By reducing variation in techniques we reduce variation in the process time required to complete the task,

I camcorded a unit being processed through each workstation by multiple operators. In order to do this I had to set up the station with a vision system (see monitor below) as most workstations utilise a traditional micro-scope with two eyepieces, whereby only the operator can actually see the detail being performed. This set up allowed me to view the detail of each of the elements of the process step, before completing a comparison of techniques used by each operator & ensuring the technique is adequately documented in the associated work instruction. In order to reach agreement in standardising some of the techniques, Manufacturing & Quality Engineering functions needed to work with operators to decide which of the techniques used should be made the standard.