Warehouse order processing and dispatching

Faulding Healthcare, Adelaide, South Australia

Moderators: Mohan Krishnamoorthy (CSIRO) and Lou Caccetta (Curtin University of Technology)

Background

From the Company's beginnings as a pharmacy in colonial South Australia in 1845, Faulding has grown to a top 100 company listed on the Australian Stock Exchange.

Faulding is a diversified world-wide health and personal care company. The Company's Principal businesses are generic oral and injectable pharmaceuticals, consumer health products, the provision of distribution and retail management services to pharmacies, and logistics management services to hospitals.

Faulding Healthcare remains the market leader in community pharmacy distribution with the largest market share

Faulding Hospital Pharmaceuticals commands the greatest market share and broadest range of generic anti-cancer medicine products in a number of countries in which it operates

Faulding Oral Pharmaceuticals is among the top 10 companies in the USA oral generic industry and the Company's record in obtaining approvals for new drugs places Faulding among the leaders in the industry

Faulding Healthcare is an integrated business unit comprising both health care services and consumer products. Faulding gains added value through providing branded products, services and solutions directly to consumers and hospitals, or via Faulding's retail pharmacy network.

Order and delivery process

The following diagram illustrates our current warehouse order processing constraints.

Click for a larger diagram.

Orders are placed by pharmacies either twice per day with add-ons (last minute additions) or progressively all day with no pattern, this can happen up to 9-9.30 pm.

They can be transmitted by Portable Data Entry 75 %, EDI 2%, FAX 5% or phone 18%. If orders are placed by certain cut off times they meet the delivery runs. Orders placed late automatically revert to the next run. Late orders can be accepted and the scheduling system over ridden, but this is not the norm.

The normal order process analysis is then performed i.e. credit check, stock availability (if not in stock the pharmacy is notified real time through the PDE at order placement) pricing issues etc etc.

The order is then analysed for tote split (a tote is a plastic tub that the order is delivered in). The tote split is determined by the volume (square metres) or weight limits associated with lifting safety. This applies for specific areas of the warehouse, mainly the split case picking area. Order picking can be carried out in three different ways. The majority of picking is done using "man-on-board" vehicles called datamobils. These vehicles can carry up to 8 totes and move through the warehouse along a pre-defined path. Once they start their cycle through a designated area of the warehouse they are programmed to stop at positions where order items are to be picked. Another form of order picking (not included in our study) uses a machine called "Cathy" which is a fully automated crane single pick machine. Finally, some orders may be manually picked. The types of products to be picked are divided into two sections, namely, Ethical or OTC. Ethical refers to prescription drugs and OTC refers to "Over The Counter" which means any other product in a chemist shop. Ethical and OTC are in separate sections of the warehouse. A datamobil will pick only Ethical or OTC, but not both, in a given cycle.

The order can also be split into other areas i.e. bulk, where full cartons are picked, DD's where dangerous drugs of addiction are picked, Fridge, where fridge lines are picked. The mini bulk area is the stock kept immediately next to the datamobil area to be used to replenish the split case picking shelving.

Once all order processing analysis is performed (automatically on our mainframe software - "VDS") the scheduling function is then also automatically performed. This uses an in house system developed by our programmers in which orders are sorted by delivery run and 8 totes are allocated to a datamobil cycle on a first in first serve basis. The orders are manually released by a line start operator.

The order/datamobil cycle is generally released in delivery run priority. We can release any cycle if required as long as it has reached the scheduling screen. The manual releasing rationale is based on lines per cycle, operator competency, window of opportunity to meet the dispatch time etc.

The orders/8 totes cycle is infra red data transferred to a stationary datamobil. Once this is complete it then enters the picking zone. The OTC and Ethical areas are separate zones. Datamobils either enter the OTC area or the ethical area, and cannot go through both. The LVR is the datamobil program controlling, shelf location stopping, in which tote to place the product, and the product quantity and type correctness (bar code scanning or weight correctness on the individual scales) etc etc. A datamobil cannot proceed to the next pick point unless all such order properties have been checked.

When the datamobil cycle is complete the actual picking data/information is infra red transferred back into the mainframe and an invoice produced at the unload area. The invoices are manually placed into the correct totes and the totes are placed on a conveyor system and head for the dispatch dock. The datamobil then moves back to the line start area.

The invoice details are placed on a delivery run manifest, printed by dispatch and loaded accordingly.

Specific problem objectives

Clearly it is essential to meet orders within their due date as efficiently as possible. In the current automated warehouse operating in Sydney, 27 datamobils are engaged in order picking. Since each datamobil travels along a fixed path, and passing within aisles is not possible, some considerable vehicle interference occurs. Thus one possible objective would be to reduce the number of datamobils in operation, since this reduction would reduce the level of interference and also reduce capital and maintenance costs. Consistent with this objective would be a reduction in datamobil cycle times.

References

Gibson, D. & Sharp, G. 1992, `Order batching procedures', European Journal of Operational Research, vol. 58, pp. 57-67.

Axsater, S. 1993, `Exact and Approximate Evaluation of Batch-Ordering Policies for Two-Level Inventory Systems', Operations Research, vol. 41, no. 4.

Hwang, H., Baek, W. & Lee, M 1988, `Clustering algorithms for order picking in an automated storage and retrieval system', International Journal of Production Research, vol. 26, no. 2, pp. 189-201.

Burkard, R., Fruhwirth, B. & Rote, G. 1995, `Vehicle routing in an automated warehouse: Analysis and optimization', Annals of Operations Research, vol. 57, pp. 29-44.

Elsayed, E., Lee, M. & Scherer, E. 1993, `Sequencing and batching procedures for minimizing earliness and tardiness penalty of order retrievals', International Journal of Production Research, vol. 31, no. 3, pp. 727-738.

Rosenwein, M. 1996, `A comparison of heuristics for the problem of batching orders for warehouse selection', International Journal of Production Research, vol. 34, no. 3, pp. 657-664.

Pan, C-H. & Liu, S-Y 1995, `A comparative Study of Order Batching Algorithms', Omega, International Journal of Management Science, vol. 23, no. 6, pp. 691-700.