Integral Logistics Management — Operations Management and Supply Chain Management Within and Across Companies

4.2 Push Logistics and Pull Logistics in the Design of Business Processes

Intended learning outcomes: Differentiate between pull logistics and Push Logistics. Describe the temporal synchronization between use and manufacturing with inventory control processes.

4.2.1 Pull Logistics

Process transitions arise when several people or groups work independent­ly of one another within a business process. In general, for a value-adding process of any complexity, a business process must be divided into a number of subprocesses. Crucial are the states of goods between sub­processes and particularly the event (see above) that detects the state, or momentary standstill. Two subprocesses must be connected by an interface. This guarantees that the they cannot be torn apart in time, but will take place one right after the other.

Figure emphasizes the fact that the customer’s logistics are ongoing throughout this whole period. The customer keeps track of order fulfillment, as the goods ordered are needed for fulfilling the customer’s own tasks (in design and manu­facturing) or for use.

Fig.        Business process in the enterprise from order acquisition to fulfillment.

How should the business process be organized into subprocesses, as soon as more people are needed for order acquisition and fulfillment than can be incorporated into one single group? Experience has shown that each transition from subprocess to subprocess is critical. This is the reason why the design of the interfaces is so important.

Figure shows a common solution to the problem. This example has been taken from a midsized company in the metals industry. Transitions are defined by the way that an order arises or is formulated between the persons or groups of persons involved.

Fig.        Interface between subprocesses: “customer-supplier relationship with an internal order” model and pull logistics.

Design and manufacturing is viewed here as its own business process, as the sales department has issued an internal order to the design/manu­facturing departments. Here, sales is the internal customer, and design/ manufacturing the internal supplier. Sales, however, remains responsible to the customer for order fulfillment during the entire design and manu­facturing period. Through continuing coordination, or in other words, the exchange of control information, the order is eventually fulfilled. The “customer,” whether internal or external, places an order and “pulls” the logistics in such a way that the logistics produce the goods ordered for delivery. The customer remains an active monitor, at least potentially, throughout the entire delivery lead time.

This results in cascades, that is, a number of process levels in the process model. Pull logistics is the name for this system: Value-adding takes place only on customer demand or to replace a use of items. Its characteristic is that several parallel order processes arise. This means that several order managing persons concern themselves with the value-adding process simultaneously. With regard to delivery reliability rate, each customer, through coordination with the supplier, “pulls” the order on up through the process levels.

This kind of logistics ensures that nothing is “forgotten.” Parallel order management in multiple levels is in itself, of course, not value-adding. From a lean production perspective in a narrow sense, it may even be wasteful. However, from an agility perspective, this slack is necessary if logistics are to be effective in this model. The interface in the cascade model is formed mainly through the formulation of the order. Customer and supplier must reach an agreement. These negotiations represent slack, and thus unnecessa­ry expenditure, but they do result in an overall effective business process.

4.2.2 Push Logistics

An alternative solution to the design of the business process in Figure is a type of logistics that is shown in Figure — a simple sequence of subprocesses.

Fig.        Interface between subprocesses: the “simple sequence” model.

This “simple sequence” model is common and effective, as long as order management does not change and remains in the hands of the same person. This person is the supplier responsible for all subprocesses; he or she manages the executing organizational units in a central fashion, one after the other. This is the model of push logistics.

With push logistics, you push the order based on a given schedule planned in advance in the direction of the added value, without need of customer influence or a definite customer order.

If decentralized control by the executing organizational units themselves is desired, the “simple sequence” model can hardly be utilized. First, there are no indications of how states between the subprocesses might be registered so that the next subprocess will be initiated. Between subprocesses, order management must be somehow shifted from one processing facility to the next. Responsibility then lies in the hands of the organizational unit that executes that next subprocess. Second, the external customer in our example must first deal with sales and later with design and manufacturing units. But how will the customer know when these transitions occur? Misunderstandings become inevitable. For these reasons, the “simple sequence” model — although “lean” — is bound to fail. Figure shows that only careful designing of the transitions between subprocesses, that is, the interfaces, can make uninterrupted order fulfillment processing possible using push logistics.

Fig.        Interface between subprocesses: “partner relationship with overlap­ping subprocesses for handing over the order” model.

The practical example in Figure is taken from a consulting firm. In the company’s past, vendors had made agreements with customers that the executing units could not fulfill. This, of course, had a negative effect on customer satisfaction. The company recognized that during contract negotiations, and at the conclusion of the agreement itself, at least one person should take part that will actually perform the services. This type of organization ensures that nothing will be sold that cannot be produced. Conversely, the executing unit commits itself at the right point in time in direct contact with the customer.

With push logistics, it is crucial that the two part processes overlap, that is, that the next part process begins parallel to the end of the preceding part process. This link is established by having people in the organizational unit handling the first part process conduct their last task in coordination with representatives of the organizational unit that will begin the second part process. This second group takes over process management — the responsibility as supplier with regard to quality, cost, delivery, and flexibility. At the same time, the party placing the order knows its “new” business partner, and order fulfillment can be coordinated.

In this model, the organizational units of subprocesses do not stand in a customer-supplier relationship, but rather stand in a partnership. The overlap of the subprocesses is the necessary slack. It is true that more persons than actually necessary perform certain subtasks. But it is this very redundancy that ensures a smooth takeover of the order by one organizational unit from the other. The two subprocesses become sewn together, and this is what makes for an overall effective business process.

It is not necessary to play off the two models in Figures (“customer-supplier relationship with an internal order”) and (“partner relationship with overlapping sub­processes for handing over the order”) against each other. Both the multiple process levels model with its pull logistics and the flat model with its push logistics have their justifications. For fast, uninterrupted pull-through of complex value-adding processes, enough slack, or non-value-adding activity, must anyway and always be built in at process transition points.

The more employees are capable of handling “longer” processes, the faster and cheaper the process­es become. The reason is that there is less necessity for slack times and redundant work in order to join subprocesses in smooth transition. Of course, qualifying employees to do this and coordinating them in the group entails costs. From this, we can derive a guide­line for the design of process organization. Division into short sub­processes may be necessary in order to achieve certain quality demands. As soon as several people show competency in the handling of a number of related subprocesses, it is correct — with a view to reducing transition points — to make a long process out of the short sub­processes and to organize these persons into a group (see also [Ulic11]).

4.2.3 The Temporal Synchronization between Use and Manufacturing with Inventory Control Processes

Section 1.1.2 introduced temporal synchronization between supply and demand as a fundamental problem in logistics. Warehouses serve the storage of goods when manufacturing or procurement is too slow or too early. Figure shows the MEDILS notation for logistics with stocking. Depending on the point of view, or the type of order (see Figure, certain cases of inventory control processes result as shown.

Fig.        Different inventory control processes for temporal synchronization between use and manufacturing / procurement.

  1. Manufacturing / procurement takes place only upon actual custom­er demand. Storage is necessary only if order receipt is too early.
  2. Manufacturing / procurement is released based on forecast, before there is a definite customer and without need to replace items taken for use. These products are then held in inventory until required by a user. They can then be delivered immediately.
  3. Demand is filled immediately from inventory. The items taken for use will then be replaced by stock replenishment The replacing items remain in inventory for an indefinite length of time.

Cases 1 and 3 can be considered to be pull logistics. Case 2 is mostly solved by push logistics, as long as there is no definite customer. However, for each case it is clear that carrying inventory only makes sense if goods in stock will be used within a reasonably short period of time. Figure shows the example from the above section once again, this time incorporating inventory control for end products following the Case 3 point of view, that is, “inventory replenishment after consumption.”

Fig.       Pull logistics with inventory: order processing with end product inventory.

Note that “design” is missing here, because this case concerns the selling of already manufactured and stored products.

Course sections and their intended learning outcomes

  • 4.4 Characteristic Features Relevant to Planning & Control in Supply Chains

    Intended learning outcomes: Produce an overview on principle and validity of characteristics in planning & control. Explain six features in reference to customer, and item or product or product family, five features in reference to logistics and production resources, as well as seven features in reference to the production or procurement order. Describe important relationships between characteristic features and features of transcorporate logistics in supply chains.

  • 4.5 Branches, Production Types, and Concepts for Planning & Control

    Intended learning outcomes: Describe branches of industry in dependency upon characteristic features. Explain in detail production types and concepts for planning & control. Disclose selecting an appropriate branch model, production type, and concept for planning & control.

  • 4.6 Summary


  • 4.7 Keywords


  • 4.8 Scenarios and Exercises

    Intended learning outcomes: Differentiate between various concepts for planning & control within the company. Using process charts for synchronization between use and manufacturing with inventory control processes. Elaborate a basic process analysis as well as manufacturing processes in the company-internal layout.

  • 4.9 References


Print Top Down Previous Next