Emmett UK is one of the largest leek producers in the UK, supplying fresh leeks and value-added products to leading retailers. It has a very strong customer focus and a good reputation. Our project examined the relatively capital-intensive operations (compared to some other agricultural sectors) from harvesting (Emmett Lincs), through packing (Emmett UK), to plant dispatch for the retailer distribution system.
Emmett UK operates its own farms and balances seasonal supply and demand from external farms. Occasionally, it reverts to the spot market, where leeks can be traded if short-term factors, such as the weather, result in an excess of supply.
FCC work demonstrated that harvesting on the farm was driven by productivity rather than quality. Major improvements and cost reductions were made by slowing production down and focusing on the quality of the final product.
Our project looked in detail at pre-packed leeks in a 0.5 kilo retail pack and followed three phases:
• Current state mapping - to understand and measure the current performance of the value chain
• Future state vision - to define an agreed picture of what the improved supply chain should look like in the future
• Lean improvement - deploying the appropriate tools to improve operational performance
CURRENT STATE
The current state map (below) shows the flow of both information and physical products. It summarises the main physical flow processes with key performance data for quality, cost and delivery. It also details all inventory holding and transport links within the site.
The information flow for the product covers customer orders, forecasting, planning and scheduling, operations, transport management and supplier ordering. The current state map summarises the departments involved in the overall flow, the data transferred across departments and any plans or instructions given to transform or move the product.
The team developed linked spreadsheets comprising a current state map, future state map and performance summary.
The current state map allowed the team to identify the following key findings:
• The value-adding time as a percentage of the total time of the product in the chain was 0.55 per cent
• Leeks are in storage for an average of 52 hours, which represents almost 80 per cent of the throughput time
• Only 65 per cent of the product reached the end of the packing line without loss, partial loss or waste
• Our analysis of the retailer weekly forecast and daily order showed a mean forecast error of between -13 per cent and three per cent, with a maximum forecast error of between -30 per cent and 20 per cent
• Packaging stock at Emmett UK had increased recently but reduced at its suppliers.
In addition, there were some very specific issues identified with harvesting operations:
• No supplier guarantee that seed will work
• Trim “properly” in field reduces trim in factory
• Measure kilo/worker/hour in the field
• Link cost of “packing out ”back to cost of labour in the field
• Evaluate costs involved against time from seed purchase to product dispatch. Compare to buying leeks outside
• Change processing to two-docked trailers operated in farmyards. Saves one day’s stock and hence one trim
• Automate leek pushdown before trim and automate “singulation” before blade
• Apply U-shaped layout to single cell many times instead of several touches by several people. Bottleneck is weighers and feed to HFFS
• Automate whole leek trimming and presentation to packing machine (HFFS)
• Farm losses increase when held for longer in store at factory. Reduces yield and affects for several weeks.
ROOT-CAUSE ANALYSIS
The team decided to focus on the pack rate because of its importance to farm profitability.
Two data sets were collated. The first showed a pack-out range of 50-80 per cent with a mean of 67 per cent. The second had a range of 45-80 per cent with a mean of 62 per cent.
The team decided to conduct a root-cause analysis of pack-out rates. Bins were placed at each work station to collect and weigh waste, and video sampling was introduced for several days from field to retail packing. The team collated exact figures for losses and logged visual evidence of the process.
The work revealed that leek inspection and trimming were two key areas of quality loss and high labour cost. The team conducted several investigations into root causes.
• Was the production line “over-inspecting”? The inspection/trim team stood back from the workstation, and the line was allowed to run through to retail packing. The quality of the packout was unacceptable.
• On the farm, were high productivity rates reducing quality? The farm director reduced the daily target by 30 per cent. The amount delivered was 15 per cent less but pack-out rate increased from 70-85 per cent.
• Data was collected on the length of leeks entering the automatic cutter to establish the optimum setting and minimise waste. The team introduced sorting and trimming on the rig at first-cut stage by using tray spacers. This allowed the automatic cutter to be set for the two different sizes, significantly reducing losses.
FUTURE STATE
Team members developed an ideal state in which everything worked perfectly. From this they developed a future state map incorporating improvements that can be made soon.
A lean supply chain should exhibit certain key features:
• Everyone in the entire value chain should be aware of the rate of end-customer consumption
• The chain should have as little inventory as possible
• The inventory that does exist should be in the right amount, in the right form and in the right place in the chain
• There should be as few transport links as possible and with minimum distances
• Information processing should be kept as simple as possible, with pure signal and no noise in the information flows
• A lean value stream will have the shortest possible lead-time
• Changes introduced to smooth flow, eliminate inventories and reduce transport and lead-times should involve the least possible, or even zero, cost.
Several ideas were generated on how an ideal state could operate. The first idea was to create a mobile factory on the farm being harvested. Several refrigerated trailers would move between farms complete with a weighing and packing line in a compact area. The second idea discussed was packing on the harvesting rig itself. The third idea, which formed the basis for the future state, evolved as the root-cause analysis demonstrated that more accurate harvesting could significantly simplify the packing process, and deliver major quality and cost benefits.
This ideal state revolved around three principles:
• One-touch-cutting on the field, correct first time
• Rapid cooling and delivery to packing to prevent telescoping (leeks continue to grow once cut if kept at ambient temperature)
• Just In Time processing by an autonomous production line.
ACTION PLAN
The Action Plan aims to prepare extra-trimmed leeks completely in the field so that they can go straight into cooling and packing in the factory without any further preparation. This “perfect” raw material implies frequent supplies from the field rig which are hydro-cooled at the factory.
There are three generic stages to the plan:
• Continuous Improvement
- One-touch-cutting - through regularly-paced cutting
- Improved factory/farm communication through a capacity-balance spreadsheet
- ‘Just In Time’ delivery
• Farm Step Changes
- New rig design
- Cooling relocated to factory
• Factory Automation
- Handling of trays
- Sorting and weighing
- Improved packing lines
* * * * * * *
Ten steps were agreed to take forward the Action Plan.
Improvement Initiative with little investment:
1. Improved manual cutting in the field
2 Determine preferred Emmett specification
3. Operations reduce labour, return ‘wrong’ leeks to field
4. Stop weighing trays on rig
5. Revise design of ‘super’ rig
6. Install auto-cutter on rig to achieve final length
Later projects with investment:
7. Automated Factory
8. Final overview
9. Masterclass
10. Achieve wider ‘lean’ capability
A lot of direct costs are labour in the harvest and packing processes. Our team estimated that almost 40 per cent of direct cost savings are linked to a very small amount of investment in continuous improvement to the process through implementation of one-touch-cutting and just in time.
The improvements these changes will bring about can be seen in the Key Performance Indicators identified to assess the impact of this project.
The Food Chain Centre wishes to acknowledge the assistance of David Simons (Cardiff Business School) and Ian Sadler (on secondment from Victoria University, Australia) for their help in preparing this case study.
No comments yet