|This paper traces the growth of the concept of inbound logistics management, with a focus on computer systems, from its
origins in the mid-1980s to its near-maturity in the late-1990s. The need for this paper is driven by both the lack of
a clear understanding of what constitutes inbound logistics, and more importantly, the recent emphasis placed on automated
inbound logistics systems as "the new guy on the block" way of dramatically reducing a firm's transportation
costs. This paper also offers speculation on likely new directions in inbound logistics for beyond Year 2000. While inbound
logistics computer systems are discussed, this paper is intended to be a non-technical discussion.|
|Inbound logistics does not have an agreed-upon definition in the industry. Several questions arise when one attempts to define
the term, for example: Where is the "in" in inbound? Is outbound the opposite of inbound?|
|Clearly the definition of inbound logistics is a matter of perspective. If you are the receiver of a shipment, the shipment is
inbound to you; if you initiate or send a shipment as a raw materials supplier, manufacturer or vendor might do
then it is outbound from you. For inbound logistics planning, the focus of transportation management is on planning the
receipt of the shipment. The definition of inbound logistics, then, is related not only to the controller of
the shipment process, but also to who takes ownership of the goods shipped.|
|Inbound logistics traffic departments do not simply manage shipments from their own companies (outbound), they plan shipments
from other companies "in" to their location. Suddenly the transportation process becomes infinitely more complex.
With this complexity comes more opportunity to reduce transportation costs if transportation is well-managed.|
|The deregulation of carrier rates is often cited as ushering in a new era of transportation management practices and a need
for automated logistics management systems. Two other events in the mid-'80s, however, more likely explain the dramatic
changes that took place in traffic departments, including:|
|Increasingly, companies/traffic departments were taking control of the "inbound" flow of goods and materials.
In other words, they were beginning to manage the flow of shipments from suppliers and vendors. The benefits of taking
control are increased visibility of these shipments (for tracking and potential redirection) and potential cost savings
due to volume shipments or better negotiated rates with carriers.|
|The "profit center" mentality had reached corporate traffic departments, with an emphasis on cutting transportation
costs. Traffic departments were tapped to reduce shipping costs and reduce transportation costs charged by suppliers and
vendors for the delivery of goods. This often meant unit-level pricing without transportation costs included. On the surface,
the profit center approach seemed to have a positive impact on the "new" inbound logistics traffic departments,
but the traffic planning workload more than doubled with the onset of corporate downsizing.|
|In conjunction with new directions in traffic departments was the emergence of software systems specifically designed to
support new logistics planning goals. In the early '80s, this simply meant software that performed carrier routing and
rate calculations for "low cost" carrier selection.|
|Another related change during this era was a resizing of computers for running transportation management software. Long
before manufacturing and inventory management systems moved from high-end mainframe systems, transportation had moved to
mid-range and PC-based hardware. This afforded transportation several advantages, including:|
- Local control of systems and databases, eliminating the shared environment.
- The cost advantage of less expensive equipment.
- The ability to run software more efficiently on the PC.
|These key events of the 1980s began to transform traffic departments into more independent (but dependent upon) information
systems departments. Traffic also became more integrated with accounting/finance due to a combined goal in reducing corporate
|The dramatic changes that occurred in transportation departments in the mid-'80s included:|
- Traffic departments felt increasing pressure to mesh with the mid-'80s corporate culture of cost savings/income
generation strategies. Traffic was asked to be a "center of excellence."
- Corporate assessment of traffic became more performance-oriented with threats of replacement with third-party logistics
- While corporations sought cost justification and profit center results, they also wanted to minimize capital investments
required by traffic departments. Traffic departments were "caught between a rock and a hard place," with a major
need for new software and hardware for new inbound logistics management activities.
|By the mid-'80s, a significant change had occurred in logistics systems. A birth occurred in systems which attempted to
optimize transportation (i.e., minimize overall transportation costs against optimizing criteria or rules). Such
software followed two paths: linear programming solutions, which combined orders to be shipped in all possible combinations
to achieve a minimum cost; and heuristic models, which, against a set of logistically valid rules, asked the same question,
but did not test all combinations of the orders. The definition of each approach is covered elsewhere; here we note that this
new software sought to minimize the transportation costs across several modes of movement: truckload (TL), less-than-truckload
(LTL), rail, multi-modal, etc. In general, this new software sought to turn LTL moves into TL moves involving multi-stop
moves that took advantage of cheaper per-mile TL rates.|
|By 1985, transportation software systems included more than simple routing, minimum path, traveling salesman solutions. Although
several software vendors played in the industry at that time, only a few began to offer solutions that could optimize loads and
put optimal orders together to create multi-stop shipments.|
|Pioneers in multi-modal load optimization planning included: Distribution Sciences, Inc., Optimal Decisions Systems, Inc. and
United Management Consultants. Some of these vendors no longer exist; some have been absorbed by other players in the software
industry. With a focus on carrier (vs. shipper) planning from/to pool locations, Mentis Corporation and Sunflower Systems
(eventually Weseley; then McHugh Software; now RedPrarie) planned multi-stops in and out of pool-points.|
|What prevailed in the load planning systems was "outbound" shipment systems, involving shipments from a single point
(manufacturing plants, warehouses, DCs and pool points) to many points (customers, stores). For the most part, through the
early '80s any discussion of inbound logistics management was a mere speculation.|
|Through the last part of the 1980s, software vendors of "consolidation/load optimization" software attempted to reverse
the outbound process to create an inbound logistics management system. Other attempts to build inbound logistics planning
software with multi-stops assigned a "seed" order or location and attached nearby orders, often according to weight
or volume. Such solutions almost always resulted in non-minimum path solutions, and therefore higher costs. Additionally,
large orders (a.k.a., "greedy" solutions) begot smaller orders, ultimately leaving smaller orders
|Again, these solutions failed to address the way inbound logistics planning worked, i.e., shipments involved multiple
stops at suppliers or vendors to single or multiple destinations points which were distribution centers or depots. Reversing
the outbound algorithms failed to address the complexity of the inbound process.|
|The mid- to late-'80s defined the following relative to inbound logistics management:|
- Traffic departments realized that monies were lost by not controlling the inbound logistics process.
- Corporations were less interested in building transportation costs into product prices; corporations wanted to address all
areas of price cutting separately.
- A focus on inbound logistics planning increased, assuming better rates could be negotiated with carriers and better service
could be achieved with increased volume of traffic.
- Inbound logistics reduced the non-controllable LTL moves and was assumed to minimize handling damage and address the touted
just-in-time (JIT) delivery of goods. Most importantly, inbound logistics management addressed on-time delivery requirements.
- For the retailer, inbound logistics planning addressed a need to minimize inventory levels, improve turns and downsize DC
- Inbound planners attempted to "beat" the transportation costs proposed by suppliers and vendors with the option
to select whom plans the move.
- As a result of watching inventory levels, increasingly shipments (purchase orders) involved LTL shipments. This required
combining shipments from multiple vendors/suppliers into full truckloads.
- An increased need existed for "visibility" of shipments. For retailers, this visibility insured that shipments
reached stores to meet ad dates; for manufacturers, visibility enabled them to redirect shipments relative to MRP planning,
i.e., to meet all components of supply chain management.
|As the inbound logistics planner moved into the 1990s, other issues became part of the transportation planning equation:|
- Pre-ticketed, store destination goods began moving through a cross-dock facility or directly to the final destination.
As a result, the complexity of inbound moves required more advanced algorithms in transportation software to process the
exponentially increasing number of alternative combinations of potential pick-up and drop-off points.
- Strong corporate desire to minimize capital investment in DCs and warehouses developed. Firms were downsizing the number
of facilities owned, (looking for software to help analyze alternatives) while placing increased importance on minimizing
network mileage in the shipping process. An increasing use of pool facilities developed as an alternative to private DCs
- Customer service became key to logistics strategies, with on-time delivery, regional warehouses to better serve customer
needs. With a focus on on-time delivery, the demands on logistics software were redefined.
|By the end of the 1980s, an intense interest was identified in inbound logistics planning. This was fueled by the Department
of Defense (DOD), which acted as its own third-party logistics planner, as well as by retailers who saw that a major reduction
in transportation costs could be achieved by taking control of the inbound planning process. The approach by both was quite
simple: convert LTL shipments to TL shipments by planning multi-stop moves.|
|After a review of inbound logistics software for DOD's Military Traffic Management Command (MTMC) in 1985, ARCLOGIX found
that vendors of optimizing load planning systems did not provide or address inbound logistics planning.|
|In the late '80s the Defense Logistics Agency (DLA) was charged with managing inbound, truckload (greater than 10,000 lbs.)
freight from DOD suppliers. As in the non-government world, approximately 80 percent of this freight was handled via
motor carrier. The DLA's Guaranteed Traffic Program, which promoted high annual volume of traffic by traffic lane and
carrier, allowed multi-stop moves. Except for some transfers between depots (DOD's DCs), shipments were from government
suppliers located throughout the United States, to a military installation.|
|At this time, DLA was experimenting with continuous move/freight consolidation (multi-stop programs out of its Boston,
Cleveland and St. Louis regional offices). After several years of testing, proprietary programs were abandoned largely due
to the unfeasible nature of manual shipping and because of the dynamically changing locations of suppliers.|
|As part of ARCLOGIX's effort to develop a carrier selection process for DLA, DLA identified that it needed a freight
consolidation, TL multi-stop move automation of its shipping process. DLA had determined that a fifteen percent reduction
in motor freight costs would result if an automated freight consolidation program were in place. The DLA narrowed its list
of vendors providing freight consolidation software in 1986, and charged ARCLOGIX with the task of building an inbound freight
consolidation system that could manage multiple stops at suppliers for pick-up and multiple stops at government facilities for
|In addressing DLA's requirement for a freight consolidation, inbound planning system, ARCLOGIX took several approaches
first developing a pooling analysis software, then creating a linehaul, multi-stop system. The pool analysis system
proved cost-ineffective in early testing for DLA because DLA did not focus on pool carrier/pool sites. For DLA the pooling
algorithm was abandoned.|
|The key requirement that drove the design for the DLA was the need to create a system providing inbound planning from multiple
suppliers (pick-up) to multiple DOD destinations. ARCLOGIX examined DLA's approach, assessed other vendors' multi-stop
load planning software and decided to take "a clean slate" approach to solving the inbound logistics system
|At this time (1987), Rick Murphy, Schneider International's former VP of information systems (and former president of
Optimal Decision Systems) consulted with ARCLOGIX to define ARCLOGIX's freight consolidation process. Rick defined the
design of outbound freight consolidation systems, which ARCLOGIX used to guide the design of its multiple-vendor origin to
multiple-destination system. To accomplish a true inbound logistics system, the ARCLOGIX software assumed that there were no
true origin-destination points in the network of pick-up and delivery locations. This assumption would not only allow for a
minimum path branch and "traveling salesman" solution to truckload routing, but would also erase the binding
requirement of dealing with an inbound versus an outbound system. The assumption would open continued modifications of the
software to accommodate post Year 2000 enhancements.|
|From 1989 to 1993, ARCLOGIX developed an inbound logistics system using new algorithms (as opposed to reversing outbound
algorithms). The first full inbound commercial installation was in 1993 at a discount retailer. Several conditions were
- The system was PC-based although in the past corporate traffic departments believed that a PC could not support the
requirements of planning 200-400 shipments per day.
- Similar software failed to plan multiple vendor or supplier pick-up and merely reversed the outbound planning process.
- Most systems at this time often took twelve hours to process a normal day's worth of orders; the ARCLOGIX software
|The software proved most applicable to the retail environment with the following a direct result of the inbound logistics
- Increased visibility of goods shipped with tracking, redirecting or expediting of shipments possible.
- Provided a cost evaluation to determine if shipments should be vendor/supplier planned or if the inbound traffic department
should arrange the shipment.
- Allowed negotiated rates with carriers by traffic lanes and true selection of low-cost carrier.
- Permitted maximum utilization of available commercial fleet during a driver strike. In fact, an ARCLOGIX retail customer
used the software to plan multiple vendor stops to multiple divisional DCs.
- Integrated a cross-dock separation into the distribution process; utilized assembly points and consolidators as the first
leg of a linehaul move.
|The above examples are from the first four installations of the ARCLOGIX inbound logistics software installed during late-1993
and early-1995 at a regional discount retailer, a national food distributor, a regional discount retailer and a national off-price
retailer. The most advanced of these application is shown in the following "Retail-Inbound Decisions" diagram.|
|The diagram shows modes of shipments made during the inbound shipment process for retailers:
- From vendor to a pool or assembly point as an LTL move; from the pool to a DC as a TL move.
- From vendor as a direct, non-multi-stop LTL move.
- As a multi-stop from two vendors with a destination stop at both the DC and a store.
- As a multi-stop from two vendors.
- As a multi-stop from the DC to several stores.
- As a backhaul shipment.
|Much has changed in the routing capabilities of inbound logistics systems since 1993. The types of multi-stop moves found
in the ARCLOGIX software, as well offered by other vendors, are defined in the box below. Evident is that inbound traffic
management has become increasingly complex, requiring an automated system to evaluate all of the possible shipping/cost
alternatives associated with moving an order from a vendor.|
- Multi-stop linehaul through a set of vendors or suppliers.
- Multi-stop linehaul including a stop at a pool point or consolidator.
- Backhaul from a vendor or supplier at the end of multi-stop linehaul.
- Continuous move of a truckload carrier involving emptied trailers at each stop.
- Multi-stop linehaul with partial unloading, loading to capacity, partial unloading in order to maintain a continuously
- Multi-stop into a pool or consolidator site in conjunction with multi-stop linehaul.
- Multi-stop linehaul pick-up from vendors/suppliers to multi-stop drop-off.
- Multi-stop into and out of pool or consolidator sites in conjunction with multi-stop linehaul.
- Multi-stop assessment of best point to ship (e.g., DC, consolidator or store/drop/ship.)
|Corporate America's traffic departments have nearly perfected the outbound planning process to the point of creating
a near "template" of the shipment process. What varies is new logistics strategies requiring modifications to
existing automated systems. Often the issue is one of managing large amounts of data with a small traffic staff.
Corporate transportation, as a consequence of conquering the outbound process, has given increased attention to managing
the inbound flow of goods and materials as the next cost-cutting or strategy to enhance transportation management.|
|In the second half of the 1990s, inbound logistics management of traffic has been the last frontier of control (both
cost and strategic) over the transportation management process. With retailers, often impacted by minimal mark-up on
goods, reducing transportation costs proves even more important. Recent industry reports indicate that the greatest
cost savings are to be gained by automating the inbound process.|
|Here, only one change in the inbound distribution channels is identified post-Year 2000: Today's (mid-90s) distribution
process finds the more classic supply chain components manufacturers to warehouses to retail DCs to stores to customers,
as seen in the following diagram by A. St. Onge as printed in Supply Chain Management Review, Spring 1997:|
|St. Onge has identified that shipments will increasingly go directly from the manufacturer to the customer, bypassing the
retailer's distribution center and stores (note his 7% today versus 25% in 2005 diagram). St. Onge uses the move
towards consumer purchases from catalogs as evidence of this trend. ARCLOGIX extends the St. Onge observation to note
a general trend in inbound shipping practices (see diagrams - ARCLOGIX estimates in brackets):|
|Retailers will merge, or reduce capital investments in general in distribution centers.|
|Direct store, from vendor, will become more important, but will require receiving characteristics (scheduling; unloading
dock capabilities) and inventory storage conditions for each store.|
|Direct customer delivery from the manufacturer will increase, and is present today in outbound moves, as example, direct
delivery of materials to building sites.|
|Increasing use of cross-dock facilities, pool sites or consolidators to replace distribution centers or warehouses, with
direct store or customer delivery from these sites. This logistics strategy has required pre-ticketing or orders identified
by store. For example, one of ARCLOGIX's retailer customers is closing all but one of its DCs and using pool carriers
to move all freight into stores, another retailer is converting to all cross-dock DCs.|
|In all cases, the inbound logistics software requirements for optimization routines will greatly extend the possible
combinations that an order can be routed to minimize the overall transportation costs. The key here, with past inbound
systems the cost minimization among the vendors was considered. In 2005 systems, the software will be required to optimize
both at the origin and at the destination of shipments, i.e., to consider which is most cost effective shipping
to a DC, store or customer site.|
|The new inbound system then breaks down the barrier between inbound and outbound logistics management. In fact, the maturity
of requirements for the inbound system overshadows the need for an outbound system. The inbound system of post Year 2000
will consider both inbound and outbound moves as if they are independent of their respective origins or destinations.
Redefining all shipments as inbound results in a focus on customer service, i.e., concentration on the delivery date
rather than the destination. This also conforms to the now old concept of just-in-time delivery.|
|While this paper was written several years ago, the forecasts on the direction of TMS software continues to be true today.
But history, even for TMS software, does tend to repeat itself. Look for an upcoming paper from ARCLOGIX on how the concepts
associated with artificial intelligence have surfaced to drive the "new" TMS software.|
|Lisa Harrington, "The Best Offense is a Good Defense," Inbound Logistics, January 1995|
Arthur St. Onge, "Personalized Production: Riding the Information Superhighway Back to the Future,"
Supply Chain Management Review, Vol. 1, Spring 1997