Optimisation of warehouse management in a fulfilment centre

Optimisation of the Warehouse Management

 

Having collated the findings, we devised several initiatives to optimize the warehouse management in the fulfilment centre, improve performance, and, ultimately, retain the client’s customers.

Optimisation of client area distribution:
Using our Multi-Layer Path Model, we were able to perform a cost-benefit analysis of various warehouse layouts. This allowed us to arrange and optimise the distribution of the client areas in an informed manner. We consolidated clients into single areas or multiple areas next to each other so that warehouse operatives would no longer have to cross the warehouse for orders from a single client. We also placed important, high-frequency clients beside the picking access areas so order picking runs would be shorter for more frequently accessed clients.

Improved compartment selection:
We categorised products as slow-moving or fast-moving. We then introduced a regime whereby fast-moving items would be placed on the mid shelves, where picking is quickest, and the slow-moving items would be placed at the top or the lowest level, where picking takes significantly longer. Item size also became a determining factor for compartment selection: small items, such as headphones and textiles, would no longer be placed in large compartments, where they failed to fill up the space. Additionally, items that were often ordered together were placed in compartments next or near to each other to reduce the required movements of the pickers.

Improved re-stocking processes:
We devised new settings for the re-stocking software that would recommend the optimum compartment to re-stock to. The respective logic is shown in the picture below. New compartments were used only where necessary, otherwise additional quantities were added to already existing compartments. With improved confidence in the software, we were able to oblige the re-stocking operatives to follow its recommendations – no more overwriting of the suggested compartments. Additionally, the pickers had to prioritise picking from the emptiest compartments. This acted as a 'self-cleansing' system, clearing out the half-filled compartments and freeing up warehouse space.

 

 

Improved re-stocking parameters:
Based on historical data, we calculated two things: the re-order point for each item and the re-order quantity for each item. As a result, compartments for slow-moving items did not become needlessly over-filled, while for fast-moving items, there was now enough time to replenish the stock before a stock outage occurred. In addition, more than three boxes were res-stocked for fast-moving items, so that replenishment was needed no more than once a day.

Introduction of ‘batch’ picking runs:
Whereas previously picking runs would operate continuously and be routed by ‘gut feeling’ we created a ‘batch’ regime, whereby particular runs would take place every 1-5 hours, depending on requirements. Due to respective adjustments in the software, orders accumulated so that visits to specific sections of the warehouse would be worthwhile, resulting in filled carts and, overall, shorter picking runs.

Trolley-size optimisation:
To make the most of the new ‘batch’ order picking runs, we adjusted some of the picking trolleys from 9 compartments to 12 compartments. This was based on the order characteristics of the customer. For example, the second largest customer was a B2C business, which had smaller orders and was therefore better suited to trolleys with more separate compartments. This meant more orders could be picked on a single run, improving the efficiency of a picking run.

Hardware optimisation:
What might seem a trivial point, but one that in practice caused endless frustration and delays, was the operation of the hardware, such as printers and scanners, which were previously overburdened and slow. Updates to the software and APIs made the hardware run more efficiently, improving label and documentation printing times.

Intervals for integration of processes:
Having mapped out the entire warehouse process, we were able to determine all the points on the critical path to an order shipment. This included specific timings, such as when orders were loaded onto the system, when restocking started, and when picking runs were formed and executed. Combining this information with contractual SLAs and carrier pick-up times, we devised a new routine that coordinated these processes and timings in intervals, resulting in shorter waiting times and higher efficiency. A major enabler of this coordination was the change to ‘batch’ picking runs. This provided singular points of alignment, computation, and planning, instead of continuous (and often inaccurate) calculation and recalculation.

Optimised resource scheduling:
We introduced new work schedules to address the peaks and troughs throughout the day. If there was less to do, employee start times would be staggered, so more operatives would be on duty for the shipment cut-off time and its aftermath. Flexitime and time accounts allowed the client to address seasonal peaks with internal staff. Additionally, a new set of vacation rules – for example, a maximum proportion of operatives on leave at any one time before Christmas – meant that the fulfilment centre wouldn’t find itself suddenly shortstaffed in a time of need.

 

---

 

Results of the improved warehouse management

 

Implementation of the above initiatives resulted in significant improvements in efficiency, the highlights of which are shown below.

• The new warehouse layout led to a 17% reduction in picking times
• The intervalisation of the restocking processes led to a 20% reduction in picking warehouse stock-outs
• At the beginning of the project, the % of on-time positions was just 15%. By the end of the project, on-time positions had improved fourfold to 60%
• Items delivered ‘over 3 days late’ reduced from 55% to 2,5%

 

These efficiencies, among others, resulted in over 400k EUR in annual savings, broken down as follows:

• Through warehouse configuration optimisation (such as client area optimisation), we realised 120K EUR in annual saving
• Through process efficiency improvements (such as picker routing, loading time reduction and hardware improvements), we realised 127k EUR in annual savings
• Through compartment allocation and replenishment improvements, we realised 190k EUR in annual savings

 

We also devised further initiatives for the client related to one of its key pain points: price determination. With its newly improved performance, the client could begin to restore its reputation and increase pricing back towards the industry standard. But it wasn’t just a case of asking for more money.

Firstly, the client had to demonstrate to its customers that its performance had improved. We recommended a number of new data management measures and enhanced KPIs so the client could prove that it was meeting, or even exceeding, its contractual SLAs.

Secondly, we recommended a data-driven price negotiation regime. Rather than entering the negotiations with their backs to the wall and submitting to the customer’s demands, the client could enter negotiations with evidence-based assertions that would justify its prices.

We estimated the resultant annual revenue increase to be over 1 million EUR.